Method for feeding in electrical power by means of a wind energy system

Abstract

The invention relates to a method for feeding electrical power (P) into a power supply network (120) having a grid voltage (U) at a grid connection point (118) by means of a converter-controlled feeder (100; 112; 130), in particular a wind energy system (100; 112) and / or a memory (133). The invention also relates to a wind energy system for carrying out such a method. The invention also relates to a memory for carrying out such a method. And the invention relates to a converter-controlled feeder for carrying out such a method.

Description

Technical Field

[0001] The invention relates to a method for feeding electrical power into a power supply network. The invention also relates to a wind energy system for carrying out such a method. The invention also relates to a memory for carrying out such a method. And the invention relates to a converter-controlled feeder for carrying out such a method. Background Art

[0002] Wind energy installations are known which feed electrical power into the power supply grid. This can be done via a single wind energy installation or via a wind farm in which a plurality of wind energy installations are combined and fed into the power supply grid via the same grid connection point. Wind energy installations or wind farms having a plurality of wind energy installations are summarized under the general term wind energy system. Such wind energy systems can also have electrical storage devices, from which the power supply grid is also fed.

[0003] In this case, the feed-in takes place by means of a frequency converter or a current converter. Such wind power installations or wind parks and electrical stores which feed into the power supply grid by means of a current converter are therefore referred to as current converter-controlled feeders. In addition, photovoltaic installations can also be current converter-controlled feeders.

[0004] Nowadays, such converter-controlled feeders, in particular wind energy systems, have an increasing share in providing electrical power in power supply networks, at least in some power supply networks. Thus, such converter-controlled feeders not only form energy suppliers or power suppliers, but they also serve to electrically support the power supply network and generally assume increasing importance in this regard.

[0005] In this context, such converter-controlled feeders, in particular modern wind power installations and thus modern wind parks, must have a property for proactively controlling grid faults, which is also referred to as an FRT property.

[0006] The FRT characteristics of modern wind power plants are known and established. In this case, additional voltage-supporting reactive current is fed in during voltage dips in the power grid, in particular by means of a parameterizable droop. For this purpose, two modes are known in principle, Mode 1 and Mode 2, which can also be referred to as QU(2) mode or QU(UK) mode.

[0007] In both modes, in the event of a fault, the power is increased up to the apparent current limit of the converter. If the apparent current limit is reached, reactive current is fed in preferentially, i.e. in mode 1, reactive current is fed in preferentially, or active current is fed in preferentially, i.e. in mode 2, active current is fed in preferentially.

[0008] As is known, for voltage-flexible power grids, reactive current priority is provided, i.e. in mode 1, in order to generate, i.e. feed in, a completely reactive current. This means that the feeding in of reactive current takes precedence over the feeding in of active current, and when the reactive current is to be increased further, the active current is to be reduced, if necessary even to 0. Thus, the active current is reduced, whereby the reactive current can still increase while complying with the predefined maximum apparent current. In this case, the known relationship between apparent power S, active power P and reactive power Q is based on the following formula:

[0009] S 2 =P 2 +Q 2

[0010] Of course, the same relationship also applies to the apparent current on the one hand and to the active current and reactive current or active current component and reactive current component on the other hand. Due to the square relationship, i.e. square addition, a decrease in the active component may only lead to a relatively small increase in the reactive component, depending on the operating point. In particular, when the active component should already be small, a further decrease only leads to a minimal increase in the reactive component.

[0011] In frequency-flexible power grids, frequency problems can occur in the event of a voltage sag due to the priority of the reactive current. In this case, the voltage sag causes the priority of the reactive current to cause a reduction in the active current, thereby causing the fed-in active power to be reduced to zero at certain times. However, the reduction in the fed-in active power can cause a reduction in the frequency. Therefore, in frequency-flexible power grids, this can cause a strong frequency reduction, resulting in frequency problems.

[0012] In the case of active current priority, i.e. according to method 2, the active current feed-in has priority and, if necessary, the reactive power share must be reduced or not allowed to increase, so that when the active current can be increased as a result, no reactive power may be fed in at all. As a result, the above-mentioned frequency problem does not occur directly, but here, due to the priority of the active current in the case of a voltage sag, no voltage support may occur at all, or even a reduction in the reactive power that was already fed in before the fault may occur. Therefore, in addition, in the second mode, in the case of a voltage sag, the wind energy installation attempts to feed in the total active power via a higher current up to the current limit of the converter. Therefore, a voltage collapse is thus faced. Summary of the invention

[0013] The object of the present invention is therefore to solve at least one of the above-mentioned problems. In particular, a solution is to be proposed in which good grid support is provided in fault operation, which avoids not only the mentioned frequency problems but also the mentioned problems of potential voltage collapse. At least an alternative solution is to be proposed with respect to the solutions known to date.

[0014] According to the invention, a method is proposed for feeding electrical power into an electrical supply network.

[0015] Therefore, a method for feeding in electrical power is proposed. The feeding takes place at a grid connection, at which the feeding into the power supply grid takes place. The power supply grid has a grid voltage. For example, the grid voltage can be measured at the grid connection point. The feeding takes place by means of a converter-controlled feeder. For the feeding, such a feeder has a converter or inverter, which can thus generate the current in a targeted manner depending on the value, frequency and phase position.

[0016] Such a converter-controlled feed-in can be designed in particular as a wind energy system and can also or alternatively be designed as a store. Here, therefore, a wind farm or an individual wind energy installation can be mentioned, both of which can be understood as the term wind energy system. Converter-controlled feed-in, i.e. feed-in by means of a converter, can also particularly mean the use of a plurality of converters, in particular a plurality of converters connected in parallel.

[0017] The method proposes that, in normal operation, if no grid fault or grid disturbance is detected in the power supply network, electric power is fed into the power supply network. Correspondingly, the method performs monitoring of the power supply network for grid faults or grid disturbances. The explanation of grid faults in the following text also relates to grid disturbances in the same sense, even if the grid disturbances are not explicitly mentioned. This can be done by means of measurements and / or by means of evaluation of corresponding signals obtained from the outside, for example by evaluating signals transmitted by a grid operator of the power supply network. Preferably, such monitoring of grid faults in the power supply network can also be carried out by evaluating the performance of a converter-controlled feeder. Such converter-controlled feeders usually react to changes in the power supply network, and such reactions, which are also described below, also allow conclusions to be drawn about the power supply network or possible faults therein.

[0018] It is also possible that there is a grid fault in the power supply network, but that the grid fault is not significant or is not taken into account for other reasons, for example because the grid fault is too far away. In this respect, the feed-in during normal operation means that no grid fault is detected.

[0019] It is also proposed that if a grid fault or grid disturbance is detected, a switch to fault operation is made. Such a switch to fault operation is proposed if it is a grid fault or a disturbance in which the grid voltage decreases or increases. Thus, in particular voltage drops or voltage surges are monitored. A voltage drop is in particular a drop in the grid voltage, which drops below a predetermined lower voltage limit value of the grid voltage. In the case of voltage surges, the grid voltage rises above an upper voltage limit value.

[0020] If there is a mode in which reactive current is preferred, it is particularly meaningful to take overvoltage into account. It has been recognized that when crossing an overvoltage fault (OVRT), if a fixed active power is maintained, although the active current decreases in proportion to the voltage, the reactive current can still rise strongly, so that the limitation or reduction of the active current and thus the active power may become necessary. This is especially the case if a large amplification or amplification factor is used between the voltage deviation and the reactive power to be fed in, which is also called the k factor. Then high reactive currents can still occur, so that the active current or therefore the active power is reduced. For this reason, it is also proposed, however, not to reduce the active current to a value of zero. Undervoltage or voltage sag may also cause reactive currents with increased values ​​in order to increase the voltage. The reactive current is different from the situation in which it should act to reduce the voltage, but the reactive current is in its phase. Intuitively speaking, the reactive current is negative compared to this, which is also called "absorption". However, the problem that arises is that the maximum apparent current may be exceeded.

[0021] It is proposed that in normal operation, active current is fed in to feed in active electrical power, so that additional reactive electrical power can be fed in as required via reactive current. Active current and reactive current together result in an apparent current, in particular via I 2 S =I 2 P +I 2 Q The known relationship formed together yields the apparent current. In the formula, I S ,I P and I Q Form apparent current, active current and reactive current.

[0022] Reactive current can be fed in for voltage support, ie for voltage boosting, but also for voltage reduction with the opposite sign. However, it is also conceivable that no reactive current is fed in, at least initially.

[0023] For fault operation, it is proposed to feed in additional reactive current for increasing or reducing the grid voltage, i.e. to feed in additional reactive current with the corresponding phase. This can also be expressed in a slightly simplified manner as feed-in with different signs. The corresponding feed-in for increasing the grid voltage is also called "generation", and the corresponding feed-in for reducing the grid voltage is called "absorption". Therefore, by feeding in additional reactive current, the value of the reactive current increases.

[0024] Here, two different modes are proposed. That is, it is proposed to use a reactive power priority mode or an active power priority mode, and the reactive power priority mode is characterized in that, when necessary, the active current is reduced so that the apparent current complies with the apparent current limit, and the active power priority mode is characterized in that, when necessary, the reactive current is limited or reduced so that the apparent current complies with the apparent current limit. In order to comply with the apparent current limit, it can happen that only one of the active current and the reactive current that make up the apparent current can be increased at the expense of the other. Then, in the reactive power priority mode, the reactive current takes priority, and the reactive current is therefore increased at the expense of the active current when necessary. In the active current priority mode, it is the opposite, because the active current takes priority, so that the value of the reactive current is limited or reduced when necessary.

[0025] In the reactive power priority mode, the active current is reduced so that the apparent current complies with the apparent current limit. The concept underlying this is that while the active current increases or at least does not decrease, the reactive current additionally fed in is increased in magnitude so that the apparent current reaches its apparent current limit and also exceeds the apparent current limit until the corresponding safety protection takes effect and the current line is mechanically interrupted, or when the physical limit of the semiconductor is reached, the semiconductor protector cuts off the semiconductor. This is of course undesirable and is controlled accordingly to avoid the situation. For this purpose, the active current is reduced so that the apparent current complies with the apparent current limit, i.e. does not exceed the apparent current limit.

[0026] Alternatively, an active power priority mode is proposed. In fault operation and in a variant of the active power priority mode, the reactive current is increased to increase the grid voltage, and then the active current is increased in order to maintain the active power currently required and / or fed in until then as much as possible, because at a lower voltage more current is required to maintain the power level. However, in the case of overvoltage, in particular when more power is required, an increase in active current is also considered. Here, when the apparent current limit is reached, the reactive current is reduced so that the apparent current limit is observed when the active current increase increases further, and the apparent current limit can have the same value as the apparent current limit in the reactive power priority mode. That's all, the two modes can exist: namely a reactive power priority mode, which requires priority for reactive current; and an active power priority mode, which requires priority for active current.

[0027] In the reactive power priority mode, it is proposed that a reactive current upper limit is predefined for the value of the reactive current and, in addition or as an alternative, an active current lower limit is predefined for the active current. This allows the reactive current to be given priority in the reactive power priority mode, i.e. the active current can be reduced, so that even more reactive current can be fed in while maintaining the apparent current limit, but the limit is set so that a complete drop of the active current to zero is avoided.

[0028] In particular, it has been recognized that, due to the equations already mentioned, which describe the relationship between the apparent current, the active current and the reactive current, the reactive current can hardly be increased any more when the active current is completely reduced to zero, for example, for the last 10% with respect to the rated active current or with respect to the apparent current limit. It has therefore been recognized that, by means of the limit setting, it is possible to prevent the situation in which large damage is caused by the active current being completely reduced to zero, but which has no significant effect on the reactive current. Of course, this can also be achieved by presetting a corresponding limit value for the reactive current, in particular as an upper reactive current limit that must not be exceeded. A lower limit for the active current is preferred, because here the measurement accuracy can be higher or the setting possibility can be more precise.

[0029] According to the meaning, for the active power priority mode, it is proposed to preset an active current upper limit for the active current and / or to preset a numerical reactive current lower limit for the reactive current. Therefore, in the active power priority mode, an active current priority control can also be proposed, which can also be called an active current priority operating mode, but avoids the reactive current from falling completely to zero. Since reactive current can perform voltage support not only by forward feeding (generation), but also by forward extraction (absorption) of reactive current from the power supply network, it is proposed to preset the reactive current lower limit in terms of value.

[0030] According to one embodiment, it is proposed that in fault operation, a mode with priority to reactive power is switched to a mode with priority to active power, or vice versa. Of course, this is related to which mode exists before the switch. In particular, it is proposed that when a preset first current limit, in particular an apparent current limit, is reached, the reactive current upper limit or the active current lower limit is switched from a mode with priority to reactive power to a mode with priority to active power, or when a preset second current limit, in particular an apparent current limit, is reached, the active current upper limit or the reactive current lower limit is switched from a mode with priority to active power to a mode with priority to reactive power. In addition to the preset current limits mentioned, other limits can also be preset as the first current limit and the second current limit, for example, a current limit that is 1% to 10% lower than the apparent current limit. This makes it possible to perform the proposed switch in a timely manner.

[0031] Such a switchover is proposed in particular when riding through a network fault, for example riding through an overvoltage fault (OVRT) or riding through a low voltage fault (LVRT).

[0032] In particular, situations can occur in which the grid voltage at the grid connection point is already high, for example due to or partially due to the feed-in of active power. In this case, reactive power may have been fed in as a measure to reduce the voltage, in particular with a high amplitude, so that the reactive power feed-in approaches the maximum possible value. This may be the result of a control that wishes to adjust the mentioned high voltage downward.

[0033] Furthermore, in this case, there may be situations where a converter-controlled feeder, in particular a wind farm, feeds in an active power-first mode. In this respect, such an active power-first mode can also be set and used in normal operation, the same applies to the reactive power-first mode.

[0034] If an overvoltage fault now occurs as a grid fault to be traversed as a fault operation, the active power will remain unchanged, since it is predefined. However, due to the increased grid voltage, the active current will drop slightly. The feeder will feed in a reactive current that reduces the voltage, or increase the value of the reactive current that has already been fed in to reduce the voltage.

[0035] If the feed-in reaches the current limit, ie the apparent current limit, the reactive power or reactive current cannot be increased further.

[0036] Therefore, it is now proposed to switch from an active power priority mode to a reactive power priority mode. This is based on the knowledge that when the voltage is too high, the feeder is faced with separation from the power supply network. If this situation occurs, power can no longer be fed in at all. Therefore, in this case, compared with when the feeder remains in the active power priority mode that causes the grid separation, the switch from the active power priority mode to the reactive power priority mode causes more active power to be fed in. In other words, no active power is fed in at all. By switching to a reactive power priority operation, which however ensures that at least some active power is fed in by presetting an active current lower limit or a reactive current upper limit, at least some active power can be fed in.

[0037] In another case, it may also be advantageous to switch from the reactive power-priority mode to the active power-priority mode if the reactive power-priority mode causes the feeder to be disconnected from the grid, while the active power-priority mode prevents this. Such a switchover is particularly provided when the grid frequency is below a predeterminable switching frequency. The switching frequency is predetermined in a range below the grid rated frequency, in particular in a range below the grid rated frequency of 0.2% to 2% of the grid rated frequency.

[0038] According to one embodiment, it is proposed that at least one current limit among the reactive current upper limit, the active current lower limit, the active current upper limit and the reactive current lower limit can be changed dynamically.

[0039] The current limit prevents the reactive current or the active current, i.e. the respectively non-priority current, from being completely reduced to zero, depending on the priority. This ensures that the respectively non-priority current can still play a supporting role. However, the need for a corresponding supporting role for the non-priority current and the feasibility of providing said current can be different depending on the situation. This can be related in particular to the operating point of the converter-controlled feeder. For example, if the operating point has a relatively low apparent power, for example a relatively low apparent power consisting of a low active current component and a low reactive current component, the non-priority current component already has a low value, so that the non-priority current component can also be reduced to a lower absolute value without causing a strong negative impact.

[0040] In particular, corresponding current limits can be set according to the grid characteristics identified as characteristics of the power supply grid. Depending on the grid characteristics, the need to support the power supply grid by the remaining non-priority current share can be different. The corresponding current limits can be adapted to this. In particular, the grid rigidity further explained below can be regarded as an identified grid characteristic. The grid sensitivity or the short-circuit current ratio can also be considered as a considered identified grid characteristic, both of which are respectively related to the grid connection point.

[0041] According to one embodiment, it is proposed that, starting from the switch to the fault mode, at least one of the current limits is set or at least one of the current limits is not effective until after a predetermined waiting time. This waiting time is preferably in the range of 2 ms to 100 ms. In particular, it is in the range of 5 ms to 50 ms. The delayed effectiveness can also be achieved in the following way, that is, the relevant current limit is not taken into account until after the waiting time in the corresponding unit that is to be implemented.

[0042] It has been recognized that in the reactive power priority mode, the reactive power or reactive current is first set or regulated or controlled, and then the active power or active current is controlled or regulated, in particular by means of a superimposed controller. It has therefore been recognized that it may initially make sense not to set a lower limit for the active current, or not to allow the lower limit to take effect, or not to take the lower limit into account, because the situation, i.e., in particular, the priority conflict, has not yet occurred or is not yet of great importance. The same applies to the reactive current upper limit when a reactive current upper limit is preset in order to avoid no active current being fed in at all. During this time, the reactive power or reactive current can still be freely regulated.

[0043] According to the meaning, the concept is also proposed for the case of an active power-first mode. Here, when the active power or active current is controlled first and the reactive power or reactive current is controlled or regulated only later, in particular as a superimposed regulation, the setting or effectiveness of the corresponding current limits, i.e. the setting or effectiveness of the active current upper limit and / or the reactive current lower limit, can be stopped until then. During this time, the active power or active current can still be freely regulated or set.

[0044] Alternatively, the current limit can also be implemented as a fixed limit. In particular, a simple and stable implementation is achieved thereby. This reliably prevents the active current from being inadvertently forced to zero in a reactive power-priority mode, or the reactive current from being forced to zero in an active power-priority mode.

[0045] According to one embodiment, it is proposed that at least one of the current limits is transmitted to an external unit, in particular a grid operator. The external unit is in particular designed as an external control unit, in particular in order to control the power supply network or a grid segment of the power supply network. The external unit is therefore a unit that is not part of a converter-controlled feeder. However, the external unit can directly or indirectly pre-set, for example, a desired value for the converter-controlled feeder, wherein the external unit can also control other units in the power supply network, in particular non-converter-controlled feeders and / or switching mechanisms in the grid for connecting or disconnecting grid segments. By transmitting the information to the external unit, the external unit can in particular better perform control and / or monitoring of the power supply network, in particular better plan control and / or monitoring of the power supply network.

[0046] In particular, it is proposed that the reactive current upper limit and / or the active current lower limit are set as a function of the active power currently available for the converter-controlled feeder. This therefore relates to a reactive power-prioritized mode, i.e. a mode in which the reactive current has priority. If the converter-controlled feeder is a wind energy system, i.e. a wind energy facility or a wind farm, the reactive current upper limit or the active current lower limit can therefore be related to the available wind power. However, even in the case where the converter-controlled feeder is a memory or comprises a memory, different amounts of active power can be provided to the converter-controlled feeder, i.e. available for feeding into the power supply grid. In particular, this can be related to the charge state of the memory, so that, for example, depending on the charge state, only the associated memory power can also be provided or only allowed to be provided.

[0047] In particular, it is proposed that the lower the active power available for the converter-controlled feeder, the higher the reactive current upper limit. Correspondingly, it is proposed that the lower the active power available for the converter-controlled feeder, the lower the active current lower limit. It has been recognized that even in the case of low available active power, very high reactive currents can still be fed in. For this purpose, it has been recognized that it would seem meaningless to define the active current lower limit to a value that is higher than the active current that can be fed in due to the available active power. Correspondingly, it has also been recognized that it is also not appropriate to predefine a corresponding reactive current upper limit that actually causes a corresponding active current lower limit. It is therefore proposed to adjust the reactive current upper limit or the active current lower limit accordingly.

[0048] For example, it is proposed to define the relevant current (active current or reactive current) as a predefined proportion of the current current (or average current), for example, to define the minimum current as 90% of the current current (or average current).

[0049] It is therefore also conceivable to make the active current limit dependent on the current active power and voltage. Thus, for example, in an active power-priority mode it can be provided that the active power is allowed to drop to 80%, so that the active current limit is therefore calculated from the voltage and active power limits. Alternatively, it can be provided that the active current limit is calculated as 80% of the current active current in order to maintain the example.

[0050] For this embodiment, and also for all other embodiments involving a reactive power-priority mode, it is proposed that the lower active current limit is a maximum of 30% of the maximum apparent current. It is particularly recognized here that an active current of 30% of the apparent current is still a significantly high current, but it hardly limits the reactive current. In this case, the reactive current is limited to approximately 95% of the maximum apparent current. Thus, the maximum reactive current is reduced by 5%, but still up to 30% of the maximum active current can be fed in. Correspondingly, it is proposed that if an upper reactive current limit is predefined instead of an active current lower limit, the reactive current upper limit is at least 95% of the maximum apparent current.

[0051] However, in any case, it is proposed to set both the active current lower limit and the reactive current lower limit to a value greater than zero, and to set the reactive current upper limit and the active current upper limit to a value below 100% of the maximum apparent current. Of course, only the limits that also apply need to be set respectively.

[0052] According to one embodiment, it is proposed that in fault operation, a reactive current droop is preset in a reactive power-priority mode, wherein the reactive current droop presets the reactive current according to the grid voltage, wherein the reactive current droop describes a linear relationship between the grid voltage and the reactive current, with a reactive current increase, which represents the relationship of the reactive current change to the associated grid voltage change. However, in particular, a dead band range can be proposed, which thus describes a voltage range in which the reactive current does not increase, in particular remains at the value zero. Therefore, outside the dead band range, the reactive current increases proportionally with an increase in voltage, or decreases with a decrease in voltage. As a result, a voltage increase or a voltage decrease caused by the reactive current can be offset.

[0053] In this case, the reactive current increases in value at most up to an upper reactive current limit. The reactive current therefore remains below the maximum apparent current in value. The reactive current droop is therefore limited to the reactive current upper limit.

[0054] In addition or as an alternative to the embodiment, it is proposed that in order to comply with the maximum permissible apparent current, the fed-in active current is reduced. That is, if the increase in reactive current caused by the reactive current droop causes the apparent current obtained together with the active current to reach the apparent current limit, the active current is reduced so that a higher reactive current can be achieved for the reactive current droop. However, it is proposed that the active current is not reduced to below the active current lower limit. In this regard, priority is given to performing reactive current droop, but only to the extent that the active current does not drop below the active current lower limit. Ideally, the reactive current then reaches the reactive current upper limit. However, it is sufficient to preset the reactive current upper limit or the active current lower limit. However, two limits can also be preset, and when preset accordingly, the two limits should be reached simultaneously.

[0055] For this purpose, it is further proposed that in normal operation a reactive current droop with a reactive current increase is also provided, and that the reactive current increase in fault operation is greater than the reactive current increase in normal operation. Therefore, a steeper reactive current droop is provided for fault operation, which also takes precedence over the active current, but for which the active current is not reduced to zero. It has been recognized that, by means of a steeper reactive current droop, corresponding voltage events can be counteracted in particular by limiting the reactive current droop, i.e. by means of an upper reactive current limit or an active current lower limit, but avoiding a reduction of the active current to zero, and thus the voltage support is not significantly affected by the reactive current droop.

[0056] Preferably, a lower upper reactive current limit or a higher lower active current limit is provided for normal operation than in fault operation.

[0057] According to one embodiment, it is proposed to determine the grid rigidity as a grid characteristic for a supply grid, a subgrid or a local grid segment, wherein the grid rigidity is a measure of how strongly the grid frequency changes as a reaction to a change in the power balance in the supply grid. The power balance is therefore the relationship between the power fed into the supply grid, the subgrid or the local grid segment and the power extracted therefrom. The grid rigidity therefore describes how strongly the grid frequency reacts to a shift in the relationship between the power fed in and the power extracted, in particular a shift out of balance of the power extracted therefrom with respect to the power fed in. The less the frequency reacts to such a change in the power balance, the greater the rigidity of the grid. The grid rigidity may also be referred to as frequency rigidity.

[0058] Grid rigidity can be expressed in particular as the quotient of grid frequency change and power balance change.

[0059] For this purpose, it is now proposed that at least one current limit, i.e., an upper reactive current limit, a lower active current limit, an upper active current limit and / or a lower reactive current limit, is set according to the rigidity of the power grid. In particular, it is proposed that the higher the upper reactive current limit or the upper active current limit is set, the greater the rigidity of the power grid, or the lower active current limit or the lower reactive current limit is set, the greater the rigidity of the power grid. Here, the knowledge based on this is that there is a higher converter share in frequency-flexible power grids, so that in frequency-flexible power grids, reducing the mentioned lower current limit or increasing the mentioned upper current limit respectively causes a particularly strong priority, whereby as a countermeasure, the currents that are not prioritized respectively still have only a very small share. If a fault operation now occurs, there is the risk that the currents that are not prioritized respectively can only still perform their tasks very weakly. If the converter share in the corresponding power supply network is high, the problem occurs in a very large number of feeders, so that the problem has a high share, which may cause correspondingly large problems as described at the beginning. In order to reduce this problem, it is therefore proposed in frequency-flexible power networks to set the current limit in such a way that the highest possible proportion of the non-preferred current is maintained.

[0060] Preferably, the power supply network or subgrid or local grid section is classified into a flexible grid or a rigid grid according to the grid rigidity, and at least one current limit is set according to the classification. In this way, priorities can generally be set for the selection of current limits, and in particular when such a classification is carried out for a plurality of converter-controlled feeders of the same power supply network, subgrid or grid section and the corresponding current limit selection is carried out, a better planning safety can be achieved by the classification. In particular, it is proposed that the division and the resulting selection of the current limit are notified to an external unit, in particular a grid operator. In general, it is proposed for each embodiment that the respectively selected current limit, i.e. the reactive current upper limit, the active current lower limit, the active current upper limit and / or the reactive current lower limit, is notified to an external unit, in particular a grid operator.

[0061] In addition or alternatively, it is proposed that at least one current limit is predefined as a function of the share of converter-controlled feeders of the power supply network. For this purpose, it has already been explained above that the share of converter-controlled feeders can have an influence on the performance and thus the characteristics of the power supply network as a whole, and that accordingly different selections of current limits can be meaningful. In any case, this has been identified and it is proposed that at least one current limit is set as a function of the share of converter-controlled feeders.

[0062] In addition or alternatively, it is proposed that the grid rigidity is determined from the proportion of converter-controlled feeders of the power supply network. This has the advantage, in particular, that measurements for determining the grid rigidity may be complex and, instead, the proportion of converter-controlled feeders is essentially known. In particular, this information can be easily obtained and processed if the converter-controlled feeders provide corresponding information to a central unit of the power supply network, subnetwork or network segment.

[0063] However, it is also conceivable that the power target value is transmitted to the converter-controlled feeder on the part of the central unit and it is assumed that the share of the converter-controlled feeder can be determined depending on the presetting of the converter-controlled feeder. In particular, it is proposed for the described embodiment and all other corresponding embodiments that the share of the converter-controlled feeder is defined as the quotient of the electrical power fed into the power supply network from the converter-controlled feeder and the total active power fed into the power supply network. Alternatively, the power fed in by the converter-controlled feeder can also be proportional to the total power that can be fed into the power supply network. All these definitions apply in the same sense to the subgrids or local grid sections mentioned.

[0064] According to one embodiment, it is proposed that at least one current limit is selected as a function of the local frequency flexibility at the grid connection point and that a reactive current increase of a reactive current droop is additionally or alternatively provided. Local frequency flexibility is present at the grid connection point if the grid frequency oscillates with a greater amplitude at the grid connection point than at a reference point of the supply network. This is particularly true if the grid frequency oscillates at the grid connection point with an amplitude at least twice as great as at the reference point.

[0065] If local voltage flexibility is present in a grid section of the power supply network and a small proportion of feeders of the feed voltage are present in the grid section, local frequency flexibility is also present, and this can be an independent or supplementary definition.

[0066] In this case, synchronous generators which are directly coupled to the grid section are also considered as feeders of the feed voltage.

[0067] In particular, local voltage flexibility is present if the percentage frequency change of the power fed into the grid segment relative to the resulting percentage voltage change of the grid voltage at the reference point in the grid segment is below a predetermined comparison value for local voltage flexibility. In particular, a value of 10 or less is predefined as the predetermined comparison value for local voltage flexibility. Thus, at a value of 10, local voltage flexibility is present if the power fed into the grid segment is doubled (i.e. increased by 100%) and, as a result, the grid voltage at the reference point in the grid segment increases by at least 10% (100% / 10%=10).

[0068] Local voltage flexibility is also present, in particular, if the network segment has a relatively high impedance.

[0069] A person skilled in the art distinguishes between voltage-feeding and current-feeding feeders. In voltage-feeding feeders, the important control objective is to maintain or predetermine the voltage, i.e. the grid voltage or feed-in voltage, during the feed-in. In current-feeding feeders, the important control objective is to feed in a predefined current. Synchronous generators directly coupled to the supply grid behave as voltage-feeding feeders due to the physical properties of the synchronous generators, whereas converter-controlled feeders, in particular wind energy systems or PV installations, typically behave as current-feeding.

[0070] If the ratio of the power fed in overall by the feeders of the feed voltage to the power fed in overall by all feeders into the grid section lies below a reference ratio, in particular less than 0.2, in particular less than 0.1, then a small proportion of feeders of the feed voltage is present at the grid section.

[0071] Thus, there is a phenomenon or situation in which the power supply grid as a whole can be grid-rigid, but in grid sections, i.e. in grid sections to which the grid connection point is also connected, there are frequency-flexible grids, i.e. frequency-flexible grid sections. Such locally frequency-flexible grid sections, referred to here as locally frequency-flexible, can result in that the grid sections have a high proportion of converter-controlled feeders.

[0072] It is therefore proposed to select at least one current limit and / or the reactive current increase as a function of the local frequency flexibility in order to thereby adjust the control capability of the converter-controlled feeder.

[0073] The ratio of the amplitude of the vibration of the grid frequency at the grid connection point to the amplitude of the vibration of the grid frequency at the reference point can form a measure of the local frequency flexibility. It is preferably proposed that at least one current limit is determined according to the measure of the local frequency flexibility, and in addition or alternatively a reactive current increase is specified. This allows the control capability of the converter-controlled feeder to be set accordingly. In particular, it is proposed that the greater the measure of the local frequency flexibility, the lower the reactive current upper limit is selected, so as to thereby also ensure a correspondingly sufficient control capability of the active power feed-in. It is therefore appropriate that, in addition or alternatively, the greater the measure of the local frequency flexibility, the higher the active current lower limit can be set.

[0074] Since a frequency-flexible grid section can also be accompanied by a voltage-flexible grid section, according to one variant, it is proposed that the greater the magnitude of the local frequency flexibility, the lower the reactive current regulation capability is limited in an active current-priority operating mode. Correspondingly, the greater the magnitude of the local frequency flexibility, the lower the active current upper limit is set, or the higher the reactive current lower limit is set.

[0075] According to one embodiment, it is proposed that, in particular in the event of a fault, a reactive current droop, in particular a reactive current increase of the reactive current droop, is predefined as a function of the grid rigidity. In particular, the absolute grid rigidity already defined above is taken into account here. It is recognized that the reactive current droop can be adjusted in accordance with the grid rigidity in order to achieve a suitable reactive current control. The reactive current control can be adjusted in particular by a reactive current increase of the reactive current droop. However, other changes to the reactive current droop are also conceivable, in particular a change in the deadband range, which can therefore also be changed as a function of the grid rigidity.

[0076] The reactive current increase of the reactive current droop can be predefined depending on the local frequency flexibility, in particular depending on the dimension of the local frequency flexibility, so that the reactive current control performance of the converter-controlled feeder can be directly adapted to the conditions of the power supply grid.

[0077] In addition or alternatively, it is proposed that a reactive current droop, in particular a reactive current increase of the reactive current droop, is predefined as a function of the share of converter-controlled feeders of the power supply network. The share of converter-controlled feeders, which is specifically defined as the ratio of the active power currently fed in by all converters to the total fed-in active power, allows conclusions to be drawn about the performance of the power supply network, so that reactive current control can be predefined accordingly by correspondingly setting the reactive current droop.

[0078] In particular, it is proposed that the higher the share of converter-controlled feeders, the higher the reactive current increase in value. This is based on the recognition that the more dominant converter-controlled feeders are, the stronger the voltage support by the converter-controlled feeders should be by means of reactive current droop. More precisely, it is assumed that in the case of a small share of converter-controlled feeders, the voltage support is always provided by other units, in particular by conventional power plants with synchronous generators directly coupled to the power supply network. Basically, then, a weak reactive current control is sufficient to support the reactive current droop of the converter-controlled feeders that need to be supported.

[0079] In particular, it is proposed that a reactive current droop is predefined in fault operation. It is recognized that, in particular in fault operation, a corresponding coordination of the control of the reactive power feed-in by means of the converter-controlled feeder is important. It is also recognized that the converter-controlled feeder has a very fast control capability that can be used here. As a result, a special type of synergy with the control can be achieved in particular in fault operation. That is, a reactive current droop can be set for voltage support, possibly with a very steep reactive current increase, while maintaining a good control capability by means of the proposed current limits, without the priority of the reactive current hampering the control by the active current, or vice versa.

[0080] According to one embodiment, it is proposed to determine the converter share as the share of the converter-controlled feeder for the power supply network or a subsection of the power supply network. It has already been explained above that it has been recognized that information about the properties of the power supply network or the subsection can be given and can thus form the basis for setting the converter-controlled feeder used.

[0081] The converter share for a power supply network or a subsection can represent the ratio of the currently fed-in active power of all converter-controlled feeders to the currently fed-in active power. As a result, the current situation in the power supply network can always be taken into account with regard to the converter share in the power supply network.

[0082] Alternatively, the converter share for the power supply network or subsection can represent the ratio of the active power that can be fed in by all converter-controlled feeders to the active power that can be fed in by all feeders. In this case, the rated power of the corresponding feeder can be used as the active power that can be fed in, so that the converter share thus represents the ratio of the sum of the rated powers of all converter-controlled feeders to the sum of the rated powers of all feeders. Although considering the converter share as a ratio of the power that can be fed in may have the disadvantage of not accurately reflecting the current situation, it is a fixed quantity and is therefore also easy to determine. If necessary, if a correspondingly small power of the converter-controlled feeders can be expected due to the prevailing wind and / or the prevailing solar irradiation, an adjustment factor can be taken into account. It should be taken into account here, and it is also recognized here, that converter-controlled feeders increasingly also have electrical storage capacity. In this case, although the active power feed is currently small, the power that can be fed in can be particularly important for support measures and because the electrical storage is also - at least temporarily - callable and available.

[0083] According to the converter share, it is proposed that at least one current limit is only predefined if the converter share exceeds a predetermined minimum converter share and the minimum converter share is in particular at least 50%. It has been recognized that in the case of small converter shares, in particular when the converter share is below 50%, a loss of active power control capability in the reactive power-first mode of operation or a loss of reactive power control capability in the active power-first mode of operation can be accepted, because the remaining feeders are therefore above 50% and can mostly compensate or at least absorb the missing control capability of the converter-controlled feeders due to their physical properties, in particular when they each have a synchronous generator directly coupled to the power supply network. It has therefore been recognized that in this case, the respective full control capability of the converter-controlled feeders can be provided, in other words a full reactive current control capability up to the apparent current limit in the reactive power-first mode of operation and a full active current control capability up to the apparent current limit in the active current-first mode of operation.

[0084] According to the invention, a converter-controlled feeder is also proposed. Such a converter-controlled feeder is in particular designed as a wind energy system, i.e. as a wind energy facility or as a wind farm, or as a storage device. It is also conceivable to combine a wind energy system with a storage device. Such a converter-controlled feeder, i.e. a converter-controlled feeder according to each of the implementation possibilities described, is configured to feed electrical power into a power supply network having a grid voltage at a grid connection point. The converter-controlled feeder comprises:

[0085] a feed-in unit for feeding electrical power into the power supply network in normal operation if no network fault is detected in the power supply network,

[0086] a feed-in control device for switching to a fault mode in which the grid voltage is reduced if a grid fault or a disturbance is detected, wherein the converter-controlled feeder, in particular the feed-in control device, is configured to:

[0087] - In normal operation

[0088] - feed in active current to feed in active electrical power, and

[0089] - if necessary, additionally feeding in reactive electrical power via a reactive current, wherein the active current and the reactive current together give the apparent current, and

[0090] - During fault operation

[0091] - Feeding in or increasing the value of additional reactive current to increase or decrease the grid voltage (U N ),in

[0092] - use reactive power priority mode or active power priority mode, and

[0093] - The reactive power priority mode is characterized by:

[0094] - reduce the active current when necessary so that the apparent current complies with the apparent current limits, and

[0095] - The active power priority mode is characterized by:

[0096] - limit or reduce reactive current when necessary so that the apparent current complies with the apparent current limits,

[0097] in

[0098] - In reactive power priority mode

[0099] - preset reactive current upper limit for reactive current value and / or

[0100] - preset the lower limit of active current for active current, or

[0101] - In active power priority mode

[0102] - preset upper limit for active current and / or

[0103] - For reactive current preset value reactive current lower limit.

[0104] The feed unit can be formed in particular by one or more converters or inverters. In particular, a plurality of inverters can be fed via a DC voltage intermediate circuit, from which they are fed into the power supply grid, in particular as a parallel circuit.

[0105] The feed-in control controls the feed-in unit and can optionally also control other elements, such as the overall operation of the wind energy installation or the storage of electrical energy or the transfer of electrical energy from a storage device. The storage device can be designed in particular as an electrical storage device, in particular as a battery storage device. However, the control functions can also be distributed over a plurality of elements, so that, for example, the feed-in control controls only the feed-in unit and controls the remaining elements of the converter-controlled feeder via the operating control.

[0106] The feed control device can in particular also be switched between normal operation and fault operation. This switchover can be carried out in particular in a software-controlled manner by applying corresponding control programs or control rules for normal operation on the one hand and for fault operation on the other hand. Considering different current limits for reactive current and / or active current, whether as upper or lower limits, is easily possible and also common for known converters or inverters.

[0107] Preferably, the feed-in unit can be designed as an inverter controlled via the tolerance band method. In this case, a tolerance band is predefined for the initial current, the resulting initial current is measured, and the semiconductor switches of the inverter are controlled accordingly, depending on where the initial current lies in the predefined tolerance band. In this case, the initial current is detected in conjunction with the initial voltage so that there is always knowledge of the current currently fed in, including the active current component and the reactive current component. Accordingly, the active current and the reactive current can be predefined and controlled by correspondingly predefined tolerance bands.

[0108] All of the above-mentioned possibilities and advantages of the method for feeding in electrical power can be realized by means of such a converter-controlled feeder.

[0109] In particular, it is proposed that the converter-controlled feeder, in particular its feed-in control device, is configured to implement a method according to one of the above-mentioned embodiments. To this end, the method can be stored, for example, in each case as a control program. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] The invention is now explained in more detail below by way of example on the basis of embodiments with reference to the accompanying drawings.

[0111] Figure 1 A perspective view of a wind energy installation is shown.

[0112] Figure 2 A schematic diagram of a wind farm is shown.

[0113] Figure 3 A diagram is shown with the power profile, the voltage profile and the current profile in the event of a fault according to the prior art.

[0114] Figure 4 A diagram is shown with the power profile, the voltage profile and the current profile according to the proposed variant.

[0115] Figure 5 Different operating points for the feed current are illustrated in the complex plane. DETAILED DESCRIPTION

[0116] Figure 1 The schematic diagram of a wind power plant according to the invention is shown. The wind power plant 100 has a tower 102 and a nacelle 104 on the tower 102. An aerodynamic rotor 106 with three rotor blades 108 and a fairing 110 is provided at the nacelle 104. The aerodynamic rotor 106 is set in rotational motion by the wind during operation of the wind power plant and thus also rotates the electrodynamic rotor or rotating part of the generator, which is directly or indirectly coupled to the aerodynamic rotor 106. The generator is arranged in the nacelle 104 and generates electrical energy. The pitch angle of the rotor blades 108 can be changed by a pitch motor at the rotor blade root 109 of the corresponding rotor blade 108.

[0117] The wind energy installation 100 has a generator 101, which is shown in a pod 104. Electric power can be generated by means of the generator 101. A feed-in unit 105 is provided for feeding in electric power, which feed-in unit can be designed in particular as an inverter. In this way, a three-phase feed-in current can be generated according to amplitude, frequency and phase for feeding in at the grid connection point PCC. This can be done directly or also together with other wind energy installations in the wind farm. In order to control the wind energy installation 100 and also the feed-in unit 105, an installation control device 103 is provided. The installation control device 103 can also obtain preset values ​​from the outside, in particular from a central power plant computer.

[0118] Figure 2A wind farm 112 is shown with three wind energy installations 100 by way of example, which can be identical or different. The three wind energy installations 100 therefore represent in principle any number of wind energy installations of a wind farm 112. The wind energy installations 100 provide their power, i.e. in particular the current generated, via a power plant grid 114. In this case, the current or power generated by the individual wind energy installations 100 is added, and a transformer 116 is usually provided, which steps up the voltage in the power plant so that it can then be fed into a power supply grid 120 at a feed-in point 118, also usually referred to as a PCC. Figure 2 This is only a simplified view of a wind farm 112, which does not show a control device, although it is of course present. The power plant grid 114 can also be designed differently, for example, by having a transformer at the output of each wind energy installation 100, just to cite another exemplary embodiment.

[0119] The wind farm 112 also has a central plant computer 122 which can be connected to the wind power plant 100 via a data line 124 or wirelessly in order to exchange data with the wind power plant and in particular to obtain measured values ​​from the wind power plant 100 and to transmit control values ​​to the wind power plant 100 .

[0120] Figure 2 A memory 130 is also shown, which can also be operated together with the wind energy installation 100 or can also be used alone as a feeder for converter control. For this purpose, the memory 130 has a feed-in unit 135, which can be controlled via a memory control device 133. The memory control device 133 can be connected to a central power plant computer in order to be controlled via this. The memory 130 with the feed-in unit 135 and the memory control device 133 can also be operated independently, without the wind energy installation 100 and without the central power plant computer 122, and for example fed into the power supply network 120 at its own grid connection point. The feed-in unit 135 of the memory 130 can be constructed like the feed-in unit 105 of the wind energy installation 100 and can be operated accordingly.

[0121] Figure 3 A diagram with three individual diagrams is shown. The upper diagram shows the profile of active power P and reactive power Q. The middle profile shows the profile of the grid voltage for all three grid phases u1, u2 and u3. Correspondingly, the lowermost diagram shows the profile of the fed-in three-phase current with the three phase currents i1, i2 and i3. In the lower diagram, a time axis is plotted as abscissa with the time in ms. The time axis applies to all three sub-diagrams.

[0122] In the middle diagram, the voltages are plotted in volts, the absolute amplitude being less critical. What is more important is the change in amplitude. Thus, a voltage drop can be seen at a time of approximately 63 ms, in particular in the case of the first phase voltage u1 and the third phase voltage u3. The second phase voltage u2 is approximately zero at this time, so that the voltage drop is not noticeable there at this time. Therefore, starting from this time of approximately 63 ms, i.e. the start of the voltage drop, the amplitudes of the three voltages u1, u2 and u3 are reduced, and in this respect Figure 3 Based on simulations, the grid voltage is predefined in the form of three phase voltages u1, u2 and u3. Figure 3 It can be seen in the middle diagram of that the voltage or the amplitude of the three individual voltages drops by approximately 50% starting from the time point of 63 ms.

[0123] The fed-in active power P and the fed-in reactive power Q are depicted in the upper graph in units of kW or kvar. Therefore, at the voltage drop at 63 ms, the active power P also immediately drops accordingly, which is depicted as the occurrence of a fault 302. Before the occurrence of the fault 302, there is a situation where no reactive power Q is fed in. Then, after about 5 ms from the occurrence of the fault 302 until the voltage support, the increase in reactive current feed-in begins, i.e., from zero in the case shown. This is depicted as the beginning of the reactive current increase 304. The active current is first maintained at a constant level here, i.e., its amplitude does not change. Then, at about 74 ms, the reactive current Q increases, so that the apparent current composed of the active current and the reactive current in a known manner has reached a maximum value. This is in Figure 3 Reaching current limit 306 is depicted in the upper graph of .

[0124] However, the reactive power Q fed in should be increased further, and for this purpose, since the maximum value of the apparent current must not be exceeded, the active power P decreases, to be more precise, to zero. Then, at approximately 77 ms, the reactive power Q has already reached its maximum value. This is described as the maximum reactive power 308. Then, the active power P reaches zero. Therefore, the reactive current corresponds in terms of its amplitude to the maximum apparent current.

[0125] In the lower diagram, the current curve can be seen for this purpose. Before the fault occurs, that is, before 63 ms, it can be seen that the three phase currents i1, i2 and i3 have no phase shift relative to the three phase voltages u1, u2 and u3. At the latest at the time of 68 ms, that is, when the reactive current increase 304 begins, the current transitions into the transition region. At the time of 74 ms, that is, at the time of reaching the current limit 306, it is clearly visible at least in the lower diagram that the first phase current i1 has reached the maximum value. From the time of 77 ms, that is, when the maximum reactive power is reached at the reference numeral 308, the transition region of the three phase currents ends, and it can be seen that there is a phase shift of approximately 90° between the three phase currents i1, i2 and i3 on the one hand and the three phase voltages u1, u2 and u3 on the other hand. That is, only reactive power is fed in.

[0126] The problem here is therefore that the active power P drops completely to zero and can no longer make a contribution. The active power can neither make a power contribution to the supply nor can it play a supporting role, in particular it can no longer play a frequency supporting role.

[0127] Figure 4 In this regard, improvement suggestions are shown and Figure 4 The chart is shown in the form of a diagram. Figure 3 Based on the chart of Figure 4 In FIG. 1 , a fault occurs at time t=0 and a voltage drop at voltage u1 can be seen in particular here. In parallel, a jump in the voltage phase can also be seen here. The active power P then immediately drops. In the transition region, changing reactive power values ​​can also occur, in particular due to jumps in the phase of the external voltage.

[0128] Then, at a time point of about 15 ms, the reactive power Q is increased. The active power P is also slightly increased, which is due to the attempt to compensate the active power. Then, at a time point of about 20 ms, the current limit for the apparent current is also reached, so that the active current is reduced to further increase the reactive current. Therefore, the reactive power Q is further increased, while the active power P is reduced.

[0129] and Figure 3 Different, in Figure 4 There is also a jump in the voltage phase. This jump causes the reduction in reactive power Q shown and can also cause an additional delay in feeding in, because first the reference system, ie the frequency and in particular the phase of the grid voltage, must be found, ie the feeding must be adapted thereto.

[0130] However, it is now proposed that the active current and thus the active power should not drop completely to zero. For this purpose, in particular Figure 4It can be seen from the upper diagram of that the active power P rises slightly again and finally, at approximately 100 ms, reaches a - although lower - final value. It can be seen that, in parallel, the reactive power Q only decreases insignificantly. It can thus be achieved that the control potential by the active power P remains present and the voltage-supporting effect of the reactive power Q is still not significantly impaired.

[0131] exist Figure 4 A waiting time is also implemented in the active power control circuit so that the active power can first be reduced to 0 and then increased, ie, the active current reaches or exceeds the active current lower limit.

[0132] from Figure 4 It can also be seen in FIG. 1 that the solution proposed here makes it possible that, despite the support of the voltage by the additional reactive current, there is no permanent drop in the active power.

[0133] Figure 5 The diagram illustrates in the complex plane a method for feeding in electrical power while showing different apparent currents, which can also be referred to as total current or feed-in current. I Active current I P and reactive current I Q Apparent current I The tip of the current vector thus indicates the current operating point 502 at which the electrical power is currently fed in by the feeder controlled by the converter. In addition, the apparent current limit 504 is depicted as a semicircle. The value of the apparent current must not exceed the value of the apparent current limit 504, since otherwise the safety switch would be triggered. I The apparent current vector 506 depicted therefor may therefore reach a maximum of the semicircle of the predefined apparent current limit 504 .

[0134] Operating point 502 shows the situation in normal operation. If the feeder controlled by the converter is a wind energy facility or a wind farm, the active current I P The magnitude of depends essentially on the available wind power. For the purpose of illustration, the phase angle is chosen to be relatively large. In normal operation, especially when no voltage support is provided, the phase angle can also have the value zero. Then, the reactive current I Q The value of is also zero.

[0135] If a fault now occurs, the converter-controlled feeder switches to fault mode. Here, a grid fault or grid disturbance is assumed, which causes a reduction in the grid voltage. A reduction in the grid voltage can also be described as a voltage dip. In this case, the converter-controlled feeder should now be controlled in reverse during fault mode. The supply network should therefore be supported by the converter-controlled feeder.

[0136] For this purpose, in the reactive power-first mode, one of the two described possibilities in fault operation is shown, the reactive power and thus the reactive current is increased in value. I Q In particular, it is proposed that the increase is brought about within a reaction time of, in particular, 5 ms. Then, the reactive current I Q This is depicted in Figure 5 as a reactive current increase 508, i.e., by approximately extending the reactive current I Q or the dashed line of the reactive current vector 510 associated therewith. However, the reactive current I Q The maximum increase is possible until apparent current limit 504 is reached. Thus, first a first new operating point 512 is obtained, which therefore lies on the semicircle that describes apparent current limit 504 .

[0137] In order to be able to increase the reactive current further, the active current must be reduced. The operating point can then be moved from the first new operating point 512 along the apparent current limit 504, ie along the semicircle depicted. Figure 5 , as depicted by an additional reactive current increase 514 along the corresponding arrows of the apparent current limit 504. It can be seen that the reactive current I Q It can be further improved. Active current I P It can also be seen here that the reactive current I Q The boost and active current I P Especially for very large reactive currents whose values ​​have already reached the value of the apparent current limit. I Q , the active current must be reduced very greatly for a further small increase in the reactive current. In order to prevent this, an upper reactive current limit 516 can be provided.

[0138] If the reactive current increases to the reactive current upper limit 516, the changed apparent current vector 518 is used to obtain the changed apparent current I Until the maximum possible value of the reactive current, the reactive current difference ΔI cannot be reached due to the reactive current upper limit 516Q To this end, however, a minimum active current I Pmin , the minimum active current is different from the reactive current by ΔI Q Most times.

[0139] Therefore, a large active current component can be achieved by discarding a small reactive current component. Pmin Alternatively or additionally to the preset reactive current upper limit 516, an active current lower limit 520 can also be assumed. Therefore, if the reactive current does not exceed the reactive current upper limit 516 and / or the active current does not fall below the active current lower limit 520, the minimum active current I Pmin .

[0140] In principle, it is also possible to extract reactive current from the power supply network for voltage support. This is generally also called "absorption", while positive feed-in is called "generation". These two variants are Figure 5 In the figure, it is depicted with the abbreviation "abs" or "gen". Correspondingly, in contrast to the reactive current upper limit 516, there is a reactive current upper limit 516' for the case of reactive current extraction. Therefore, the correspondingly adjusted reactive current upper limit 516' should also not be exceeded in value.

[0141] for Figure 5 The above description relates to a reactive power-first mode, in which a reactive power or reactive current priority is thus present. The reactive current priority is however limited by an upper reactive current limit 516 or 516 ′ or by an active current lower limit 520 .

[0142] In the active power priority mode, the active power or active current takes priority. However, in the event of a fault, i.e. when switching to fault operation, the reactive current can first be increased for voltage support, i.e., for example, also starting from operating point 502, according to reactive current increase 508 to a first new operating point 512.

[0143] However, in the active power priority mode, as much active power as possible should be fed in in order to also maintain the currently required active power as much as possible. If the grid voltage decreases and therefore the voltage at which it is fed in, this will inevitably lead to a reduction in the active power while the active current remains unchanged. In order to subsequently increase the active power back to the initial value as much as possible, the active current must be increased. This is exactly what is required in the active power priority mode. Therefore, starting from the first new operating point 512, the active current is increased, while the reactive current is reduced at the same time, so that the apparent current limit 504 is not exceeded. The operating point then correspondingly starts from the first new operating point 512 along the semicircle that forms the apparent current limit 504, according to Figure 5Movement “downwards.” Correspondingly, a further active current increase 522 is depicted as a corresponding arrow in order to illustrate the change in the operating point.

[0144] However, in the active power priority mode, it is now also provided that the priority current should not rise to 100%, with the consequence that the non-priority current, here the reactive current, does not drop to zero. For this purpose, an active current upper limit 524 is preset. Thus, the active current can be increased for a long time until it reaches the active current upper limit 524. The reactive current has correspondingly decreased here and a new operating point 526 has emerged. As a result, a changed apparent current I Due to the active current upper limit 524 , the active current cannot be increased to a maximum value, ie to the value of the apparent current upper limit. Therefore, the active current difference ΔP remains, which indicates the current that cannot be fed in.

[0145] But at the same time, the reactive current only drops to the minimum reactive current I Qmin Here, the maintained or minimum reactive current I Qmin Active current difference ΔI P By limiting the active current increase to a value only slightly below the maximum value, it is still possible to ensure the possible feeding of a relatively large reactive current. This makes it possible to achieve a maximum increase in active current up to the active current upper limit, or to prevent the reactive current from falling below the minimum reactive current I Qmin , or comply with these two criteria. Therefore, the reactive current lower limit 528 limits the reduction of the reactive current in the corresponding quadrant. When the reactive current is extracted, the reactive current lower limit 528' can be set accordingly for the reactive current extraction. The value is not lower than the reactive current lower limit. The active current upper limit 524 acts unchanged in the quadrant receiving the reactive current.

[0146] Therefore, a solution is proposed, in which, in the event of a fault, when the grid voltage drops suddenly, not only a reactive power-prioritized operating mode but also an active power-prioritized operating mode can be implemented, wherein at the same time it is avoided that the respectively non-prioritized current, i.e. the active current or the reactive current, is not reduced too strongly, in particular not to zero.

[0147] In particular, so-called FRT strategies, ie strategies provided for riding through grid faults, can thereby also be improved.

[0148] The idea is in particular that if there is reactive current priority or active current priority, then in the event of a fault, a dynamic reactive current or active current limit is supplemented or a minimum active current or a minimum reactive current is specified for the reactive current priority operating mode and / or for the active current priority operating mode, and the reactive current priority operating mode and the active current priority operating mode can also be synonymously described as reactive current priority control or active current priority control.

[0149] The proposed solution can be used for wind power plants, in particular for wind power plants with FACTS properties and parameterized dynamic grid support. The solution can be used for storage and charging infrastructures with parameterized dynamic grid support. In particular, an improvement in the grid characteristics in voltage- and frequency-flexible grids should be possible. This can be achieved in particular by adjusting dynamic limits or minimum currents.

[0150] In particular, the development of a fault ride-through strategy for power grids with a high converter penetration is also proposed here. The proposed strategy for riding through power grid faults is particularly suitable here also for voltage- and frequency-flexible power grids. In some cases, stable fault detection and power grid stability can also be achieved in order to assume system responsibility via a corresponding converter-controlled feeder.

[0151] It is known that the known methods for prioritizing active and reactive current are well suited for power grids with low renewable penetration. This includes the idea that the operating modes, i.e. both the reactive power-priority mode in fault operation and the active power-priority mode in fault operation, are supplemented with dynamic reactive current or dynamic active current limits.

[0152] Alternatively or additionally, a minimum active current can also be specified with priority given to the reactive current, or a minimum reactive current can be specified with priority given to the active current.

Claims

1. A method for feeding in electrical power, the method being used for - A feeder (100; 112; 130) controlled by means of a current transformer - At the grid connection point (118) - feeding electrical power into a power supply grid (120) having a grid voltage U, The method comprises the following steps: - if no grid fault or grid disturbance is detected in the power supply grid (120), the electrical power is fed into the power supply grid (120) during normal operation, If a grid fault or grid disturbance is detected, a switch is made to a fault mode, in which the grid voltage U is reduced or increased, wherein - In normal operation as described - Feed active current I P , to feed in active electrical power P, and - If necessary, additional reactive current I Q The reactive power Q is fed in, wherein the active current I P and the reactive current I Q Together we get the apparent current I ,as well as - In the fault operation - Feeding in or increasing the value of an additional reactive current in order to increase or reduce the grid voltage U, wherein - use reactive power priority mode or active power priority mode, and - The reactive power priority mode is characterized by: - Reduce the active current I when necessary P , so that the apparent current I Comply with the apparent current limit (504), and - The active power priority mode is characterized by: -Limit or reduce the reactive current I when necessary Q , so that the apparent current I Complying with the apparent current limit (504), in - In the reactive power priority mode - For the reactive current I Q The value of the preset reactive current upper limit and / or - presetting an active current lower limit for the active current, - so as to prevent the active current from falling completely to zero, or - In the active power priority mode - For the active current I P Preset active current upper limit (524) and / or - For the reactive current I Q Preset value reactive current lower limit (528), - so as to avoid the reactive current dropping completely to zero.

2. The method according to claim 1, It is characterized in that The converter-controlled feeder (100; 112; 130) is a wind energy system and / or a storage.

3. The method according to claim 1 or 2, It is characterized in that - In the fault operation, switching from the reactive power priority mode to the active power priority mode, or vice versa.

4. The method according to claim 3, It is characterized in that - when a preset first current limit is reached, switching from the reactive power priority mode to the active power priority mode, or When a predefined second current limit is reached, switching is performed from the active power-prioritized mode to the reactive power-prioritized mode.

5. The method according to claim 4, It is characterized in that - When the apparent current limit, the reactive current upper limit or the active current lower limit is reached, switching from the reactive power-prioritized mode to the active power-prioritized mode.

6. The method according to claim 4, It is characterized in that - When the apparent current limit, the upper active current limit or the lower reactive current limit is reached, switching from the active power-priority mode to the reactive power-priority mode.

7. The method according to claim 1 or 2, It is characterized in that At least one of the following current limits is dynamically variable: - the reactive current upper limit (516), - the active current lower limit (520), - the active current upper limit (524), and - the reactive current lower limit (528), and / or - at least one of the current limits is not set or becomes effective until after a predetermined waiting time has elapsed since the switchover to the fault mode, and / or - transmitting at least one of the current limits to an external unit.

8. The method according to claim 7, It is characterized in that At least one of the following current limits is set as a function of a grid characteristic identified as a characteristic of the power supply grid (120): - the reactive current upper limit (516), - the active current lower limit (520), - the active current upper limit (524), and - the reactive current lower limit (528).

9. The method according to claim 7, It is characterized in that - transmitting at least one of the current limits to a grid operator.

10. The method according to claim 1 or 2, It is characterized in that The reactive current upper limit (516) and / or the active current lower limit (520) are set according to the active electrical power P currently available to the feeder controlled by the converter.

11. The method according to claim 1 or 2, It is characterized in that In the fault operation, a reactive current droop is predefined in the reactive power-priority mode, which predefined a reactive current I as a function of the grid voltage U Q ,in - The reactive current droop indicates that the grid voltage U and the reactive current I Q The reactive current droop has a reactive current increase, and the reactive current increase describes the ratio of the reactive current change to the corresponding grid voltage change. - and the reactive current I Q The maximum increase is up to the reactive current upper limit (516), and / or in order to comply with the maximum permissible apparent current I The active current I fed in is reduced P , where the active current I P is reduced to below the active current lower limit (520), wherein In the normal operation, a reactive current droop with a reactive current increase is also provided, and the reactive current increase is greater in the fault operation than in the normal operation.

12. The method according to claim 11, It is characterized in that - The reactive current droop outside the dead band indicates the relationship between the grid voltage U and the reactive current I Q A linear relationship exists between the reactive current droop and the reactive current increase, which describes the ratio of the reactive current change to the associated grid voltage change.

13. The method according to claim 7, It is characterized in that - for the power supply grid (120), the subgrid or for a local grid section, a grid rigidity is determined as a grid characteristic, wherein The grid rigidity is a measure that indicates how strongly the grid frequency f changes in response to changes in the power balance in the power supply grid (120), wherein The power balance describes the ratio of the electrical power fed into the power supply grid (120), the subgrid or the local grid segment to the electrical power drawn therefrom.

14. The method according to claim 13, It is characterized in that - The grid rigidity is defined as the quotient of the grid frequency change and the power balance change, where - setting at least one current limit or at least one of the current limits (516; 520; 524; 528) as a function of the grid stiffness.

15. The method according to claim 14, It is characterized in that - the power supply grid (120) or the subgrid or the local grid segment is classified into a flexible grid or a rigid grid according to the grid rigidity, and the at least one current limit is set according to the classification, and / or - presetting the at least one current limit (516; 520; 524; 528) as a function of the share of converter-controlled feeders of the power supply network (120), and / or - determining the grid rigidity as a function of the proportion of converter-controlled feeders of the power supply grid (120).

16. The method according to claim 7, It is characterized in that - According to the local frequency flexibility at the grid connection point (118) - selecting said at least one current limit (516; 520; 524; 528), and / or - the reactive current increase which provides for the reactive current droop, wherein If the grid frequency oscillates with a greater amplitude at the grid connection point (118) than at a reference point of the power supply grid (120), a local frequency flexibility is present at the grid connection point (118), wherein The ratio of the amplitude of the vibration of the grid frequency at the grid connection point (118) to the amplitude of the vibration of the grid frequency at the reference point forms a measure of the local frequency flexibility.

17. The method according to claim 16, It is characterized in that - According to the scale of the local frequency flexibility - selecting said at least one current limit (516; 520; 524; 528), and / or - an increase in the reactive current is provided.

18. The method according to claim 13, It is characterized in that - According to the grid rigidity preset and / or, - presetting of the share of the feeder controlled by the converter of the power supply network (120) - The reactive current droops.

19. The method according to claim 18, It is characterized in that - Preset in the fault operation according to the grid rigidity and / or, - the proportion of feeders controlled by the converters of the power supply network (120) is predefined in the fault mode - The reactive current droops.

20. The method according to claim 18, It is characterized in that - According to the grid rigidity preset and / or, - presetting of the share of the feeder controlled by the converter of the power supply network (120) - the reactive current increases as the reactive current drops.

21. The method according to claim 19, It is characterized in that - Preset in the fault operation according to the grid rigidity and / or, - the proportion of feeders controlled by the converters of the power supply network (120) is predefined in the fault mode - the reactive current increases as the reactive current drops.

22. The method according to claim 7, It is characterized in that - determining a converter share as a share of converter-controlled feeders for the power supply network (120) or a subsection of the power supply network (120), wherein the converter share for the power supply network or the subsection - describes the ratio of the currently fed-in active power of all converter-controlled feeders to the currently fed-in active power in total, or - describes the ratio of the active power that can be fed in via all converter-controlled feeders to the active power that can be fed in via all feeders, where At least one of the current limits (516; 520; 524; 528) is only predefined if the converter share exceeds a predetermined minimum converter share.

23. The method according to claim 22, It is characterized in that The converter share for the supply network or the subsection describes the rated power P of all converter-controlled feeders. N The ratio of the sum of the rated powers of all feeders to the sum of the rated powers of all feeders.

24. The method according to claim 22, It is characterized in that The minimum converter share is at least 50%.

25. A converter-controlled feeder (100; 112; 130) for feeding electrical power into a power supply network (120) having a grid voltage U at a grid connection point (118), the converter-controlled feeder comprising: - a feeding unit (105; 115), which is used to feed the electrical power into the power supply network (120) in normal operation if no network fault is detected in the power supply network (120), - a feed-in control device (103; 133) for switching to a fault mode if a grid fault or disturbance is detected, wherein the grid voltage is reduced, wherein the converter-controlled feed-in device is configured to: - In normal operation as described - Feed active current I P , to feed in active electrical power P, and - If necessary, additional reactive current I Q Reactive electrical power Q is fed in, wherein the active current and the reactive current together give the apparent current I ,and - In the fault operation - Feeding in or increasing the value of an additional reactive current in order to increase or reduce the grid voltage U, wherein - use reactive power priority mode or active power priority mode, and - The reactive power priority mode is characterized by: - Reduce the active current I when necessary P , so that the apparent current I Comply with the apparent current limit (504), and - The active power priority mode is characterized by: -Limit or reduce the reactive current I when necessary Q , so that the apparent current I Complying with the apparent current limit (504), in - In the reactive power priority mode - presetting a reactive current upper limit (516) for the reactive current value and / or - presetting an active current lower limit for the active current (520), - so as to prevent the active current from falling completely to zero, or - In the active power priority mode - presetting an upper limit of active current for the active current (524) and / or - a numerically predefined reactive current lower limit (528; 528') for the reactive current, - so as to avoid the reactive current dropping completely to zero.

26. The converter-controlled feeder (100; 112; 130) according to claim 25, It is characterized in that The converter-controlled feeder (100; 112; 130) is a wind energy system and / or a storage.

27. Converter controlled feeder (100; 112; 130) according to claim 25 or 26, It is characterized in that The feed control device is configured to: - In normal operation as described - Feed active current I P , to feed in active electrical power P, and - If necessary, additional reactive current I Q Feed in reactive power Q, and - In the fault operation - Feeding in or increasing the value of an additional reactive current in order to increase or reduce the grid voltage U.

28. Converter controlled feeder (100; 112; 130) according to claim 25 or 26, It is characterized in that The converter-controlled feeder is configured to carry out a method according to any one of claims 1 to 24 .

29. The converter-controlled feeder (100; 112; 130) according to claim 28, It is characterized in that The feed control device is configured to carry out the method according to any one of claims 1 to 24 .

Images