DE102022133173A1 - Method for operating a large-scale storage system, computer program product and control device for carrying out such a method, large-scale storage system with such a control device and power grid with such a large-scale storage system - Google Patents

Abstract

The invention relates to a method for operating a large storage system (3) - in particular an AC-coupled one - with a plurality of power units (7) that can be connected to a power grid (1) via a common busbar (5), wherein the power units (7) of the plurality of power units (7) each have at least one battery (11) and can be operated with a separately assigned unit power, wherein at least one distribution cycle for distributing power to the plurality of power units (7) is carried out with the following steps:
- Determining a system performance requirement for the bulk storage system (3);
- Equating a power requirement to be distributed with the system power requirement;
- determining a respective state of charge for the plurality of power units (7);
- Distributing the power requirement to be distributed to the plurality of power units (7) based on the determined charge states and a respective capacity of the power units (7), wherein an allocated power requirement is determined for each of the power units (7) of the plurality of power units (7);
- checking whether a residual power requirement not distributed in step d) is different from zero and whether the large-scale storage system (3) still has free capacity;
- if the remaining power requirement is different from zero and the large-scale storage system (3) still has free capacity, equating the power requirement to be distributed with the remaining power requirement and repeating steps d) to e);
- if the remaining power requirement is not different from zero or the large storage system (3) has no free capacity, terminating the distribution cycle and assigning the respectively allocated power requirements to the power units (7).

Description

The invention relates to a method for operating a large-scale storage system, a computer program product and a control device for carrying out such a method, a large-scale storage system with such a control device and a power grid with such a large-scale storage system.

Such a large-scale storage system has a plurality of power units, with each power unit having at least one battery. During operation of the large-scale storage system, different states of charge (SOC) typically arise for the batteries of the various power units. This is disadvantageous on the one hand because the performance - both in the charging and discharging direction - of the large-scale storage system can vary depending on the current differences in the states of charge, and on the other hand because the service life of the various batteries does not develop evenly. To solve this problem, the operation of such large-scale storage systems is sometimes interrupted in order to carry out equalizing charges. This in turn is disadvantageous because the large-scale storage system is not available while an equalizing charge is being carried out. In addition, it is possible that such equalizing charges, in particular as active equalization from one power unit to another power unit of the same large-scale storage system, may be prohibited by regulation or law if the connection of the power units via a medium-voltage level is already in the public power grid.

The invention is therefore based on the object of creating a method for operating a large-scale storage system, a computer program product and a control device for carrying out such a method, a large-scale storage system with such a control device and a power grid with such a large-scale storage system, wherein the disadvantages mentioned are at least reduced, preferably do not occur.

The object is achieved by providing the present technical teaching, in particular the teaching of the independent claims as well as the embodiments disclosed in the dependent claims and the description.

The object is achieved in particular by creating a method for operating a large-scale storage system - in particular an AC-coupled system - with a plurality of power units that can be connected to a power grid via a common busbar, wherein the power units of the plurality of power units each have at least one battery and can be operated with a separately assigned unit power, wherein at least one distribution cycle for distributing power to the plurality of power units is carried out with the following steps - in particular in the following order: a) determining a system power requirement for the large-scale storage system; b) equating a power requirement to be distributed with the system power requirement; c) determining a respective state of charge for the plurality of power units; d) distributing the power requirement to be distributed to the plurality of power units based on the determined states of charge and a respective capacity of the power units, wherein an allocated power requirement is determined for the power units of the plurality of power units; e) Checking whether a residual power requirement not distributed in step d) is different from zero and whether the large storage system still has free capacity; f) if the residual power requirement is different from zero and the large storage system still has free capacity, equating the power requirement to be distributed with the residual power requirement and repeating steps d) to e) - whereby an execution or repetition of steps d) to e) represents a run through of a power distribution loop of the method; g) if the residual power requirement is not different from zero or the large storage system has no free capacity, ending the distribution cycle and assigning the respectively allocated power requirements to the power units. The method advantageously ensures in a simple manner that the individual power units and thus also batteries are as uniformly charged as possible, whereby in particular maximum performance of the large storage system is ensured in both the charging and discharging directions. At the same time, a uniform development of the service life of the batteries is ensured, whereby power per battery is also minimized, which advantageously contributes to reduced aging of the batteries. Furthermore, there is no need for disadvantageous or impermissible equalization charges. In particular, the availability of the large-scale storage system is increased.

In the context of the present technical teaching, a large-scale storage system is understood to mean in particular a battery storage system, wherein each power unit of the large-scale storage system has in particular a nominal power of at least 30 kW, preferably more than 30 kW, and/or a capacity of at least 30 kWh, preferably more than 30 kWh. Each power unit has at least one battery, wherein each individual battery has a plurality of battery cells, in particular electrochemical cells, at for example lithium-ion cells, which are electrically connected in series and/or in parallel to one another to represent a certain voltage and/or power. In particular, the individual battery cells are connected in series with one another, in particular in groups, with such groups of battery cells in turn being connected in parallel with one another. In a preferred embodiment, such a group of battery cells, preferably lithium-ion cells, can have a voltage of 600 V. If a power unit has a plurality of batteries, the batteries of the power unit can be connected or coupled to one another, in particular via a DC busbar, i.e. a direct current busbar.

The common busbar, via which the power units of the large storage system can be interconnected or coupled to one another, is in particular an AC busbar, i.e. an alternating voltage busbar. In this way in particular, the large storage system is designed as an AC-coupled large storage system.

In particular, the large-scale storage system is an AC-coupled large-scale storage system, wherein each power unit of the plurality of power units of the large-scale storage system is a DC-coupled large-scale storage system.

The different power units of the large storage system can have the same or different nominal powers. Alternatively or additionally, different batteries of a power unit can have the same or different capacities.

In the context of the present technical teaching, a system power requirement is understood to mean in particular a power requirement by the power grid or by an operator of the power grid. The system power requirement can have a negative or positive sign, depending on whether electrical energy is to be stored in the large-scale storage system (by convention, in particular a negative sign) or withdrawn from the large-scale storage system (by convention, in particular a positive sign).

The fact that the system power requirement is determined in step a) means, in one embodiment of the method, that the system power requirement is determined, in particular by a control device of the large-scale storage system. In another embodiment of the method, the system power requirement is received, in particular by the control device of the large-scale storage system, in particular from the power grid or from the operator of the power grid. In particular, in this case the control device has an interface that is set up to receive the system power requirement from the power grid or the operator of the power grid.

In particular, in step d), the respectively allocated power requirement for the power units of the majority of power units is determined taking into account a total capacity of the majority of power units. In particular, a distribution factor is calculated for each power unit, which results as the quotient of the capacity of the respective power unit divided by the total capacity of the power units, whereby a share of the power to be distributed determined in the current run of the power distribution loop is calculated by multiplying the power requirement to be distributed by the respective distribution factor. The power requirement allocated to a respective power unit results in the current run of the power distribution loop in particular as the sum of the allocated power requirement determined in the previous run for the respective power unit and the share of the power to be distributed determined in the current run.

In particular, in step e), the remaining performance requirement is calculated as the difference between the performance requirement to be distributed and a distributed performance requirement - which is not explicitly calculated in one embodiment. In particular, in one embodiment, the remaining performance requirement is calculated implicitly in step d) by adding up non-allocated performance amounts. This advantageously does not require an explicit calculation of the distributed performance requirement.

In particular, in step e) it is checked whether the large-scale storage system still has free capacity to meet at least part of the remaining power requirement.

In particular, following step g) - in particular in step h) - the performance units are operated with the respectively allocated performance requirements as the separately assigned unit performances.

Wherever a "loop" is mentioned here and in the following, for example in relation to the power distribution loop, the use of the term "loop" is not associated with a restriction to a specific program implementation. Rather, such a loop can be implemented in one embodiment as a program loop in the narrower sense; However, in another embodiment it is also possible for such a loop to be implemented recursively or in another suitable manner.

According to a further development of the invention, it is provided that only operational power units of the majority of power units are included in the power distribution. In this way, it is advantageously ensured in particular that the system power requirement can be met as far as possible, preferably completely.

In particular, in a step c0) - in particular before determining the respective state of charge - it is checked for the power units of the majority of power units whether the respective power unit is ready for operation. In particular, only the power units that are ready for operation are then included in the power distribution loop, i.e. the power distribution loop only runs via the power units that are ready for operation. The computational effort for the method is low, in particular in comparison to an embodiment described below, in which non-operational power units are assigned capacities or power requirements with an amount of zero.

Alternatively or additionally, in step c0) or in step d), the capacity of a non-operational power unit is set to zero. In particular, in this case, all power units are included in the power distribution loop, with non-operational power units implicitly not being taken into account. In one embodiment, the operational readiness of the power units is checked in each run of the power distribution loop, in particular in step d).

In particular, by taking into account the capacity and/or operational readiness of the individual power units, it is ensured that maintenance or failure of individual power units or a component of a power unit, for example a battery, is taken into account in the power distribution - in particular automatically.

According to a further development of the invention, it is provided that - in particular in a step c1), in particular before or after determining the respective state of charge - a power limit is determined for each of the plurality of power units, wherein the allocated power requirements for the power units in step d) are additionally determined in each case as a function of the assigned power limit. In this way, various nominal powers and/or momentary or systematic limitations, including aging effects, of the individual power units can advantageously be taken into account. In particular, overloading of power units can be avoided. In particular, power units can dynamically limit their power or capacity via feedback of the power limit, for example because they are too warm or completely charged or discharged.

According to a further development of the invention, a provisional power requirement for the power units is determined based on the determined charge states and the respective capacity of the power units, wherein the provisional power requirement is compared with the assigned power limit, wherein the allocated power requirement is set equal to the provisional power requirement if the provisional power requirement does not exceed the assigned power limit, and wherein the allocated power requirement is set equal to the assigned power limit if the provisional power requirement exceeds the assigned power limit - in particular in terms of amount. This represents a particularly simple consideration of the power limit. In one embodiment, the power limit is defined as an amount, wherein the allocated power requirement is compared with the power limit in terms of amount, i.e. without taking the sign into account. In another embodiment, it is possible for a first power limit to be used for a positive power requirement, wherein a second power limit, different from the first power limit, is used for a negative power requirement. In this case, it is advantageous to take into account different behavior of the power units or batteries when storing and releasing electrical energy. The fact that the provisional power requirement exceeds the assigned - depending on the sign, first or second - power limit value also means in this case that the provisional power requirement is greater in magnitude than the assigned first or second power limit value.

According to a further development of the invention, in step d) a weighting factor is determined based on the determined charge states for the power units of the plurality of power units, wherein the allocated power requirement is determined in each case as a function of the assigned weighting factor. This represents a particularly simple possibility of taking into account different charge states of the power units and thus aligning them with one another as a result.

In particular, the respective weighting factor is determined based on the determined charge states in comparison to an average charge state of the majority of power units. In one embodiment, the average charge state is calculated as an arithmetic mean of the determined charge states. Alternatively, it is possible for the average charge state to be calculated as a geometric mean or as another average of the determined charge states.

In one embodiment, the comparison with the average state of charge is carried out before the first execution of step d). In another embodiment, the comparison with the average state of charge is carried out in each run of the power distribution loop, in particular in a step d1).

According to a further development of the invention, step d) comprises the following sub-steps for each power unit: d1) determining a differential charge level as the difference between the determined charge level of the respective power unit and the average charge level; d2) determining the distribution factor depending on the capacity of the respective power unit; d3) determining the weighting factor depending on the determined differential charge level of the respective power unit, and d4) determining the allocated power requirement for the respective power unit depending on the power requirement to be distributed, the distribution factor and the weighting factor. In particular, the power requirement to be distributed is limited by the power limit value. In this way, it can be easily ensured that the power is distributed between the respective power units depending on both the capacities and the charge levels of the respective power units.

In particular, as already explained above, the distribution factor is determined as a function of the capacity of the respective power unit in relation to the total capacity calculated as the sum of the capacities of the power units, i.e. as the quotient of the respective capacity divided by the total capacity.

According to a further development of the invention, steps a) to g) are carried out at a predetermined time interval - in particular repeatedly, in particular periodically. In particular, the power distribution can advantageously be carried out cyclically on a regular basis in this way, in particular regardless of whether the system power requirement changes. In particular, it is possible that a state of the large storage system changes, for example due to changed charge states, changed operating temperatures or changed restrictions of the power units, which may make a redistribution of even a constant system power requirement necessary or at least sensible.

Alternatively, steps a) to g) are carried out - in particular repeatedly - upon a start signal. Advantageously, the power distribution can be carried out in this way depending on demand, which can save resources and in particular computing power and/or computing time.

Alternatively, steps a) to g) are carried out - in particular repeatedly - if the system power requirement changes. Advantageously, in this case, the power distribution is only carried out if the system power requirement actually changes and there is therefore a need to redistribute the power. In other operating situations, in particular if the system power requirement remains constant, computing power and/or computing time can be saved.

The object is also achieved by creating a computer program product that includes machine-readable instructions on the basis of which a method according to the invention or a method according to one or more of the previously described embodiments is carried out when the computer program product is executed on a computing device, in particular a control device for a large storage system - in particular an AC-coupled one. In connection with the computer program product, in particular those advantages arise that were previously explained in connection with the method.

The object is also achieved by creating a control device for a large-scale storage system - in particular an AC-coupled one - wherein the control device can be operatively connected to a plurality of power units of the large-scale storage system, in particular operatively connected, and is set up to carry out a method according to the invention or a method according to one or more of the previously described embodiments. In connection with the control device, in particular those advantages arise which have already been explained previously in connection with the method or the computer program product.

In particular, the control device is operatively connectable or operatively connected to all power units of the large storage system.

In particular, the control device is a computing device on which a computer program product according to the invention or a computer program product according to one or more of the previously described embodiments is installed or runs.

The object is also achieved by creating a large storage system with a plurality of power units that can be interconnected via a common busbar, wherein the large storage system has a control device that is operatively connected to the power units and is set up to carry out a method according to the invention or a method according to one or more of the previously described embodiments. In connection with the large storage system, the advantages that have already been explained in connection with the method, the computer program product or the control device arise in particular.

The large-scale storage system is designed in particular as previously explained explicitly or implicitly in connection with the method.

In particular, the large storage system comprises a control device according to the invention or a control device according to one or more of the previously described embodiments.

The object is finally also achieved by creating a power grid that has an alternating voltage network and a large storage system according to the invention - in particular an AC-coupled one - or a large storage system - in particular an AC-coupled one - according to one or more of the previously described embodiments, wherein the large storage system is electrically connected to the alternating voltage network. In connection with the power grid, the advantages that have already been explained in connection with the method, the computer program product, the control device or the large storage system arise in particular.

• 1 eine schematische Darstellung eines Ausführungsbeispiels eines Stromnetzes mit einem Ausführungsbeispiel eines Großspeichersystems sowie einem Ausführungsbeispiel einer Steuervorrichtung, und
• 2 eine schematische Darstellung einer Ausführungsform eines Verfahrens zum Betreiben des Großspeichersystems in Form eines Flussdiagramms.

The invention is explained in more detail below with reference to the drawing.

• 1 a schematic representation of an embodiment of a power grid with an embodiment of a large storage system and an embodiment of a control device, and
• 2 a schematic representation of an embodiment of a method for operating the large-scale storage system in the form of a flow chart.

1 shows a schematic representation of an embodiment of a power grid 1 with an embodiment of a large storage system 3 and an embodiment of a control device 9.

The power grid 1 comprises in particular an alternating voltage grid 2 and the - in particular AC-coupled - large storage system 3, wherein the large storage system 3 is electrically connected to the alternating voltage grid 2.

The large storage system 3 has a plurality of power units 7 that can be connected together via a common busbar 5, as well as the control device 9, which is operatively connected to the power units 7 and is set up in particular to carry out a method described in more detail below. In particular, the common busbar 5 is an AC voltage busbar.

The power units 7 each have in particular an inverter 8 and at least one battery 11 - three batteries 11 are shown as an example - in particular a plurality of batteries 11, wherein each power unit 7 is preferably designed as a DC-coupled large storage system and can be operated in particular with a separately assigned unit power. In particular, the batteries 11 of a power unit 7 are each connected together via a DC voltage rail 10. The control device 9 is in particular operatively connected to the inverters 8 for controlling them.

2 shows a schematic representation of an embodiment of a method for operating the large storage system 3 on the power grid 1 in the form of a flow chart.

Identical and functionally equivalent elements are provided with the same reference symbols in all figures, so that reference is made to the preceding description in each case.

• Es wird a) einem ersten Schritt S 1 eine System-Leistungsanforderung Ps des Stromnetzes 1 an das Großspeichersystem 3 festgestellt; insbesondere wird die System-Leistungsanforderung Ps durch die Steuervorrichtung 9 ermittelt oder empfangen. Die System-Leistungsanforderung Ps ist vorzeichenbehaftet und weist per Konvention insbesondere ein negatives Vorzeichen auf, wenn elektrische Energie in das Großspeichersystem 3 eingespeichert werden soll (Laden), sowie ein positives Vorzeichen, wenn elektrische Energie aus dem Großspeichersystem 3 entnommen werden soll (Entladen). Dieselbe Vorzeichenkonvention wird auf alle hier betrachteten Leistungsbeträge angewendet.

As part of the method, at least one distribution cycle for distributing power to the plurality of power units 7 is carried out with the following steps:

• a) in a first step S 1, a system power requirement Ps of the power grid 1 to the large storage system 3 is determined; in particular, the system power requirement Ps is determined or received by the control device 9. The system power requirement Ps is signed and by convention has a negative sign in particular when electrical energy is to be stored in the large storage system 3 (charging), and a positive sign if electrical energy is to be taken from the large storage system 3 (discharging). The same sign convention is applied to all power amounts considered here.

In a second step S2, b) a power requirement P _a to be distributed is set equal to the system power requirement Ps. In addition, c) a respective state of charge SoC(LE _i ) is determined for the majority of power units 7. The power units 7 are divided into 2 also referred to as LE.

In one embodiment, before determining the charge states SoC(LE;), it is checked which of the power units 7 are ready for operation (step c0), whereby only the ready-to-use power units 7 are assigned an index value i. The following steps are then only carried out for the ready-to-use power units 7. In another embodiment, however, it is also possible for all power units 7 to be assigned an index value i, whereby it is also checked which of the power units 7 are ready for operation (step c0). The non-ready-to-use power units 7 are then assigned a capacity K(LE _i ) of zero, whereby they are implicitly disregarded in the subsequent power distribution.

In the second step S2, the capacity K(LE;), that is to say in particular the electrical energy storage capacity, in particular in units of kilowatt hours, is determined for each operational power unit 7 or for each power unit 7. As already explained, however, it is possible that non-operational power units 7 are assigned a capacity K(LE _i ) of zero.

For the sake of simplicity, the previously explained options of considering only operational power units 7 or alternatively all power units 7 - with a capacity K(LE _i ) of zero assigned to the non-operational power units 7 - are not explicitly mentioned below, but rather only general reference is made to "the power units 7". As explained, however, the power units 7 designated with the index i can always be either only the operational power units 7 or all power units 7, in which case the non-operational power units 7 are assigned a capacity of zero. In particular, the operational readiness of the power units 7 can be checked in each run of a power distribution loop described in more detail below.

Furthermore, in the second step S2, a power limit value P _lim (LE _i ) is determined for each power unit 7 - in particular before or after determining the respective state of charge (step c1). The power limit value P _lim (LE _i ) preferably has different values depending on the sign of the system power requirement Ps in order to take into account different behavior of the batteries 11 during charging and discharging.

Furthermore, in the second step S2, a total capacity K _total is calculated as the sum of the individual capacities K(LE _i ) of the power arrangement 7; furthermore, a remaining power requirement P _miss , an exhausted capacity K _lim and an allocated power requirement P _LE,i - the latter for all power units 7 - are set equal to zero.

The power requirement P _a to be distributed is now distributed to the majority of power units 7 based on the determined charge states SoC(LE;) and the respective capacity K(LE _i ) (step d)) by determining the allocated power requirement P _LE,i for each of the power units 7, in particular taking into account the total capacity K _ges .

This takes place in particular in the power distribution loop already mentioned, which starts with a third step S3 in which the value of the index i is initialized as a running variable, in particular set to 0. Of course, another starting value for the running variable or the index i is also conceivable, for example 1. The adjustments to the power distribution loop that may then result from this - in particular in a seventeenth step S17 - are obvious and therefore do not require any separate explanation.

Preferably, based on the determined charge states SoC(LE _i ), in particular in comparison to an average charge state SoC(LE) the plurality of power units 7, a weighting factor V; is determined for each of the power units 7 of the plurality of power units 7, wherein the allocated power requirement P _LE,i is determined in each case as a function of the assigned weighting factor V;.

For this purpose, in a fourth step S4, a differential state of charge ΔSoC(LE;) is determined as the difference between the determined state of charge SoC(LE _i ) of the power unit 7 designated by the current value of the running variable i and the average state of charge SoC(LE) calculated (step d1)), whereby the average state of charge SoC(LE) as an arithmetic mean value over all charge states SoC(LE) of the power units 7.

The comparison with the average state of charge SoC(LE), here in particular the calculation of the difference states of charge ΔSoC(LE;), can thus be carried out in each run of the power distribution loop, but alternatively also outside the power distribution loop before a first run.

In addition, in the fourth step S4, a distribution factor F; is calculated as the quotient of the capacity K(LE _i ) of the power unit 7 designated by the current value of the control variable i divided by the total capacity K _total (step d2)): $${\text{F}}_{\text{i}}=\text{K}\left({\text{LE}}_{\text{i}}\right)/{\text{K}}_{\text{total}}.$$

The calculation and use of the distribution factor F; described below has the advantageous consequence that power units 7 with higher capacity are given greater consideration in the power distribution, i.e. they are allocated more power than power units 7 with lower capacity.

In particular, a weighting factor V; is now determined as a function of the determined differential state of charge ΔSoC(LE;) of the current power unit 7 (step d3)). The calculation and use of the weighting factor V; explained here and below means that, when the system power requirement Ps is negative, the batteries 11 of those power units 7 whose states of charge SoC(LE;) are lower than the average state of charge ΔSoC(LE;) are charged more strongly, whereby the batteries 11 of those power units 7 whose states of charge SoC(LE;) are higher than the average state of charge ΔSoC(LE;) are charged less strongly. Conversely, if the system power requirement Ps is positive, the batteries 11 of those power units 7 whose charge states SoC(LE;) are higher than the average charge state ΔSoC(LE;) are discharged more strongly, while the batteries 11 of those power units 7 whose charge states SoC(LE;) are lower than the average charge state ΔSoC(LE _i ) are discharged less strongly. This advantageously leads in particular to the charge states of the power arrangement 7 being equalized overall.

In a fifth step S5, the sign of the system power requirement Ps is first evaluated to calculate the weighting factor V;. If the sign is negative (charging), the process continues in a sixth step S6. If the sign is positive (discharging), the process continues in a seventh step S7.

In the sixth step S6 (charging), the sign of the differential state of charge ΔSoC(LE;) is evaluated. If this is positive, the weighting factor V; is calculated in an eighth step S8 according to the following formula: $${\text{V}}_{\text{i}}=1-\left(\mathrm{\Delta }\text{SoC}\left({\text{LE}}_{\text{i}}\right)/100\right).$$

In this respect, it should be noted that the charge states SoC(LE;) and accordingly also the respective differential charge state ΔSoC(LE _i ) are given as percentage values with a value range from 0 to 100. Power units 7 with a higher charge state SoC(LE;) than the mean value SoC(LE) thus receive a smaller weighting factor and are subsequently charged less strongly.

If, however, the sign of the differential state of charge ΔSoC(LE;) is negative in the sixth step S6, the weighting factor V; is calculated in a ninth step S9 according to the following formula, with the absolute value function abs: $${\text{V}}_{\text{i}}=1+\left(\text{Section}\left(\mathrm{\Delta }\text{SoC}\left({\text{LE}}_{\text{i}}\right)/100\right)\right).$$

Power units 7 with lower state of charge SoC(LE;) than the average SoC(LE) thus receive a larger weighting factor and are subsequently loaded more heavily.

In the seventh step S7 (discharging), the sign of the differential state of charge ΔSoC(LE;) is also evaluated. If this is positive, the weighting factor V; is calculated in a tenth step S10 according to the following formula: $${\text{V}}_{\text{i}}=1+\left(\mathrm{\Delta }\text{SoC}\left({\text{LE}}_{\text{i}}\right)/100\right).$$

Power units 7 with higher state of charge SoC(LE;) than the average SoC(LE) thus receive a larger weighting factor and are subsequently discharged more heavily.

If, however, the sign of the differential state of charge ΔSoC(LE;) is negative in the seventh step S7, the weighting factor V; is calculated in an eleventh step S 11 according to the following formula: $${\text{V}}_{\text{i}}=1-\left(\text{Section}\left(\mathrm{\Delta }\text{SoC}\left({\text{LE}}_{\text{i}}\right)/100\right)\right).$$

Power units 7 with lower state of charge SoC(LE;) than the average SoC(LE) thus receive a smaller weighting factor and are subsequently discharged less.

If, however, the sign function (signum function sgn) returns 0 in the fifth step S5 - which means that the system power requirement Ps itself is equal to 0, i.e. the large storage system 3 should neither be charged nor discharged - the method continues in a twelfth step S12. by setting the weighting factor V; equal to 1.

In a thirteenth step S13, a new value for the allocated power requirement P _LE,i is calculated by adding the product of the power requirement P a to be distributed, the distribution factor Fi, and the weighting factor Vi to the currently stored value of the allocated power requirement P _LE,i - in the first run of the power distribution loop, this value is equal to the initial value zero from the second step S2, in each subsequent run this value corresponds to the last calculated value of the allocated power requirement P _LE,i . In this way, the allocated power requirement P _{LE ,i} is determined as a provisional power requirement for the power units 7, in particular on the basis of the determined charge states SoC(LE;) and the respective capacity K(LE _i ) of the power units 7.

• In einem vierzehnten Schritt S14 wird hierzu die in dem dreizehnten Schritt S13 ermittelte vorläufige Leistungsanforderung mit dem zugeordneten Leistungsgrenzwert P_lim(LE_i) verglichen. Ist die vorläufige Leistungsanforderung kleiner als der zugeordnete Leistungsgrenzwert P_lim(LE_i) oder erreicht diesen gerade, wird bei dem hier dargestellten Ausführungsbeispiel die zugeteilte Leistungsanforderung P_LE,i implizit gleich der vorläufigen Leistungsanforderung gesetzt, indem keine weitere Maßnahme durchgeführt wird. Überschreitet dagegen die vorläufige Leistungsanforderung den zugeordneten Leistungsgrenzwert P_iim(LE_i), wird zunächst in einem fünfzehnten Schritt S15 der Wert der Rest-Leistungsanforderung P_miss um die Differenz aus der vorläufigen Leistungsanforderung und dem zugeordneten Leistungsgrenzwert erhöht, und zugleich wird die ausgeschöpfte Kapazität K_lim um die Kapazität K(LE_i) der momentan betrachteten Leistungseinheit 7 erhöht. Sodann wird in einem sechzehnten Schritt S16 die zugeteilte Leistungsanforderung P_LE,i gleich dem zugeordneten Leistungsgrenzwert P_lim(LE_i) gesetzt. Vorteilhaft wird auf diese Weise eine Überlastung der Leistungseinheiten 7 vermieden; zugleich wird mit dem Wert der Rest-Leistungsanforderung P_miss Buch geführt über die nicht verteilten oder übrig gebliebenen Leistungsanteile, wobei über die Erhöhung der ausgeschöpften Kapazität K_lim eine Information darüber bereitgestellt wird, ob noch leistungsbereite Leistungseinheiten 7 vorhanden sind.

The allocated power requirement P _LE,i for the current power unit 7 is therefore calculated as a function of the power requirement P _a to be distributed, the distribution factor F; and the weighting factor V; (step d4)). Preferably, the power requirement P _LE,i to be distributed is also limited by the power limit value P _lim (LE _i ), as explained below:

• In a fourteenth step S14, the provisional power requirement determined in the thirteenth step S13 is compared with the assigned power limit P _lim (LE _i ). If the provisional power requirement is less than the assigned power limit P _lim (LE _i ) or just reaches it, in the embodiment shown here the allocated power requirement P _LE,i is implicitly set equal to the provisional power requirement by not taking any further measures. If, on the other hand, the provisional power requirement exceeds the assigned power limit P _iim (LE _i ), in a fifteenth step S15 the value of the remaining power requirement P _miss is first increased by the difference between the provisional power requirement and the assigned power limit, and at the same time the exhausted capacity K _lim is increased by the capacity K(LE _i ) of the power unit 7 currently being considered. Then, in a sixteenth step S16, the allocated power requirement P _LE,i is set equal to the assigned power limit value P _lim (LE _i ). This advantageously avoids overloading of the power units 7; at the same time, the value of the remaining power requirement P _miss is used to keep a record of the undistributed or remaining power shares, with information being provided about whether there are still power units 7 available that are ready to perform by increasing the exhausted capacity K _lim .

In the seventeenth step S17, the value of the running variable or index i is compared with a total number N _LE of the power units 7. This checks whether the power distribution loop has already run through all power units 7. If this is not the case, the value of the running variable i is incremented by 1 in an eighteenth step S18, and the method is continued with the new value of the running variable i in the fourth step S4.

If, on the other hand, the power distribution loop has already run through for all power units 7, a nineteenth step S19 checks whether the undistributed remaining power requirement P _miss is different from zero and whether the large storage system 3 still has free capacity to meet at least part of the remaining power requirement P _miss . The latter is checked in particular by checking whether the exhausted capacity K _lim is still smaller than the total capacity K _ges , which according to the logic explained above means nothing other than that not all power units 7 are yet being operated with the power limit value P _lim (LE _i ) assigned to them and therefore there are still power units 7 that can accept or meet additional power. This is because whenever a power duration 7 is operated with the power limit value P _lim (LE _i ) assigned to it, its capacity K(LE _i ) is added to the exhausted capacity K _lim in the fifteenth step S15. If a situation occurs in which all power units 7 are operated at their respective power limit value P _lim (LE _i ), the exhausted capacity K _lim corresponds to the sum of the individual capacities K(LE _i ) and thus to the total capacity K _total .

If both conditions checked in the nineteenth step S19 are cumulatively fulfilled, i.e. the remaining power requirement P _miss is different from zero and the large storage system 3 still has free capacity, in a twentieth step S20 the power requirement P _a to be distributed is equated with the remaining power requirement P _miss , and the power distribution loop is restarted with the third step S3 - for distributing the remaining power requirement P _miss .

If, however, in the nineteenth step S19 the remaining power requirement P _miss is not different from zero, or the large storage system 3 no longer has any free capacity, the distribution cycle is terminated, the respectively allocated power requirements P _LE,i are assigned to the power units 7, and the power units 7 are allocated to the respectively allocated power requirements P _LE,i as the separately assigned unit services.

The method as a whole is preferably carried out again - in particular repeatedly, i.e. multiple times - in a predetermined time interval, or in response to a start signal, or in the event of a change in the system power requirement Ps, i.e. started again in the first step S1 after the twenty-first step S21.

Claims (11)

Method for operating a large-scale storage system (3) - in particular an AC-coupled one - with a plurality of power units (7) that can be connected to a power grid (1) via a common busbar (5), wherein the power units (7) of the plurality of power units (7) each have at least one battery (11) and can be operated with a separately assigned unit power, wherein at least one distribution cycle for distributing power to the plurality of power units (7) is carried out with the following steps: - Determining a system power requirement for the large-scale storage system (3); - Equating a power requirement to be distributed with the system power requirement; - Determining a respective charge state for the plurality of power units (7); - Distributing the power requirement to be distributed to the plurality of power units (7) based on the determined charge states and a respective capacity of the power units (7), wherein an allocated power requirement is determined for the power units (7) of the plurality of power units (7); - Check whether a remaining power requirement not distributed in step d) is different from zero and whether the large storage system (3) still has free capacity; - if the remaining power requirement is different from zero and the large storage system (3) still has free capacity, equate the power requirement to be distributed with the remaining power requirement and repeat steps d) to e); - if the remaining power requirement is not different from zero or the large storage system (3) has no free capacity, end the distribution cycle and assign the respectively allocated power requirements to the power units (7). Procedure according to Claim 1 , wherein - only operational power units (7) of the plurality of power units (7) are included in the power distribution, wherein preferably in a step c0), in particular before determining the respective state of charge for the power units (7) of the plurality of power units (7), it is checked whether the respective power unit (7) is operational, and/or wherein - in step d) the capacity of a non-operational power unit (7) is set equal to zero. Method according to one of the preceding claims, wherein - in particular in a step c1), in particular before or after determining the respective state of charge - a power limit value is determined for each of the plurality of power units (7), wherein the allocated power requirements for the power units (7) in step d) are each additionally determined as a function of the assigned power limit value. Procedure according to Claim 3 , wherein a provisional power requirement for the power units (7) is determined on the basis of the determined charge states and the respective capacity of the power units (7), wherein the provisional power requirement is compared with the assigned power limit, wherein the allocated power requirement is set equal to the provisional power requirement if the provisional power requirement does not exceed the assigned power limit, and wherein the allocated power requirement is set equal to the assigned power limit if the provisional power requirement exceeds the assigned power limit. Method according to one of the preceding claims, wherein in step d) a weighting factor is determined for each of the power units (7) of the plurality of power units (7) on the basis of the determined charge states, in particular in comparison to an average charge state of the plurality of power units (7), wherein the allocated power requirement is determined in each case as a function of the assigned weighting factor. Method according to one of the preceding claims, wherein step d) for the power units (7) comprises the following sub-steps: d1) determining a differential charge level as the difference between the determined charge level of the respective power unit (7) and the average charge level; d2) determining a distribution factor depending on the capacity of the respective power unit (7); d3) determining the weighting factor depending on the determined differential charge level of the respective power unit (7), and d4) determining the allocated power requirement for the respective power unit (7) depending on the power requirement to be distributed, the distribution factor and the weighting factor, wherein the power requirement to be distributed is preferably limited by the power limit. Method according to one of the preceding claims, wherein steps a) to g) - are carried out in a predetermined time interval, or - in response to a start signal, or - in the event of a change in the system power requirement - in particular repeatedly. Computer program product comprising machine-readable instructions for carrying out a method according to one of the Claims 1 until 7 is carried out when the computer program product is executed on a computing device, in particular a control device (9) for a - in particular AC-coupled - large storage system (3). Control device (9) for a - in particular AC-coupled - large storage system (3), wherein the control device (9) is operatively connectable to a plurality of power units (7) of the large storage system (3) and is designed to carry out a method according to one of the Claims 1 until 7 to carry out. Large storage system (3) with a plurality of power units (7) which can be interconnected via a common busbar (5), wherein the large storage system (3) has a control device (9) which is operatively connected to the power units (7) and is set up to carry out a method according to one of the Claims 1 until 7 to carry out. Power grid (1), comprising an alternating voltage network (2) and a - in particular AC-coupled - large storage system (3) according to Claim 10 , wherein the large storage system (3) is electrically connected to the alternating voltage network (2).

Images