Method of operating a redox flow battery system

Abstract

The invention provides a method for operating a redox flow battery system comprising at least two battery modules (1) wherein the redox flow battery system comprises a measuring device (4, 5) for providing a measured variable representing a measured value of the state of charge of each battery module (1), and wherein the method comprises the steps of:-cyclically operating the redox flow battery system; -detecting a measured value by using a measuring device (4, 5); -performing at least two equalization interventions on the battery module (1) during the half cycle; and wherein the last equalization intervention performed in the half cycle is of the "overcompensation" type, and the SoC2 value for the equalization intervention is predicted by means of a pre-trained AI.

Description

Technical Field

[0001] The present invention relates to a method of operating a redox flow battery system, wherein the redox flow battery system includes a plurality of battery modules connected in series, and wherein the method involves reducing or eliminating imbalances that occur between the series-connected battery modules during charging and discharging of the battery system. Background Art

[0002] Methods for reducing or eliminating imbalances that occur between battery modules connected in series during charging and discharging of a battery system are known in the prior art. For example, DE 10 2020 108053A1 discloses such a method. Measures for reducing or eliminating the imbalance are generally referred to as “balancing”. Summary of the invention

[0003] The object of the present invention is to further optimize the methods known from the prior art.

[0004] According to the invention, this object is achieved by a method according to the independent claim. Further advantageous embodiments of the invention can be found in the dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] In the following, the present invention is explained with reference to the accompanying drawings. The accompanying drawings show in detail:

[0006] Figure 1 Battery modules;

[0007] Figure 2 Battery system;

[0008] Figure 3 Unbalanced charge / discharge cycles;

[0009] Figure 4 Balanced charge / discharge cycles via “decoupling”;

[0010] Figure 5 Balanced charge / discharge cycles through “stealing”; and

[0011] Figure 6 Each half-cycle according to the invention has several charge / discharge cycles with equalizing steps. DETAILED DESCRIPTION

[0012] Figure 1A schematic representation of a battery module is shown on the left. The battery module is designated 1. The battery module comprises a cell device designated 2, a tank device designated 3, and a measuring device for detecting controlled variables. The cell device 2 is a device having a large number of redox liquid flow cells, which can be arranged as required. For example, it can be a single cell stack, a series connection of several stacks, a parallel connection of several stacks, or a combination of series and parallel connections of several stacks. The tank device 3 serves to store electrolyte and supply electrolyte to the cell device 2. With a few exceptions, the tank device 3 comprises at least two tanks, a piping system for connecting the tanks to the cell device 2, and a pump for supplying electrolyte. Here, Figure 1 Two separate pumps are shown. The electrolyte can also be pumped with a double-head pump, ie two pumps driven by a common motor. Here, the tank device 3 is designed such that it can supply all cells of the cell arrangement 2 with electrolyte.

[0013] Figure 1 The battery module 1 shown in comprises two measuring devices for providing a measured variable, which represents a measured value of the state of charge (SoC—state of charge) of the relevant battery module. The measuring device designated as 4 is a measuring device for providing the so-called open circuit voltage (OCV). The OCV value is a measured value of the state of charge (SoC) of the battery module. The measuring device designated as 5 is a measuring device for providing the terminal voltage of the single device 2 and therefore also the battery module 1. When the battery module 1 is charged or discharged, the difference between the terminal voltage and the open circuit voltage is the voltage drop across the internal resistance of the single device 3. If the charging or discharging current and the internal resistance of the single device 2 are known, the terminal voltage represents a measured value of the state of charge of the battery module. An alternative to determining the OCV value is the so-called coulomb counting, which is also a measured value of the state of charge of the battery module. This requires a measuring device to provide the current flowing through the modules connected in series. Therefore, this measuring device for coulomb counting can also be implemented outside the battery module 1, so that the battery module 1 therefore optionally includes a measuring device for providing a controlled variable. In any case, the battery system according to the invention comprises at least one measuring device for providing a measured variable which represents a measured value of the state of charge of each battery module 1 of the battery system.

[0014] exist Figure 1 The right side of is shown a symbolic representation of the battery module 1. This symbolic representation will be used hereinafter.

[0015] Figure 2 A schematic representation of a battery system is shown. The battery system comprises at least two battery modules (one of which is designated as 1), a bidirectional power conversion system (PCS) (designated as 6), and a control device (designated as 7). The battery modules 1 are connected in series and connected to the conversion system 6. Figure 2 Four battery modules are shown, wherein the dashed lines connected in series are intended to indicate any number of additional modules. A conversion system 6 connects the battery system to the mains or a higher level electrical system. The battery modules shown connected in series are also referred to as battery strings.

[0016] In such Figure 2 In the battery system shown with identical battery modules 1, no harmful imbalances can occur. However, actual battery modules 1 differ from one another due to manufacturing differences and aging processes. In addition, different operating conditions (e.g. temperature differences) between the individual modules can also cause them to behave differently (since the internal resistance of a module depends inter alia on the temperature of the module, see below). For these reasons, actual battery modules have different efficiency values ​​and different internal resistances. For a given charging or discharging current, the higher the efficiency, the faster the relevant battery module reaches the final state. Because the same current will flow through Figure 2 All battery modules 1 in the series connection shown, so the modules with higher efficiency will reach the end state faster than the modules with lower efficiency. In order to avoid damage, the charging or discharging process must be interrupted as soon as a module reaches the corresponding end state. In this way, without compensating for this effect, the available storage capacity of such a battery system will decrease with each cycle ("capacity fade"). The different internal resistances of the modules will also have similar effects. The terminal voltage has an upper limit value that must not be exceeded and a lower limit value that must not be lowered. Even with the same efficiency, the module with higher internal resistance will reach the corresponding limit value of the terminal voltage faster during charging or discharging than the module with lower internal resistance. If the first module reaches the limit value, the corresponding process must be terminated, which also leads to a reduction in the available capacity of the battery system. Alternatively, the power of the system may also be reduced. In either case, these effects will lead to damage to the system. Balancing aims to reduce or completely eliminate the described effects in order to permanently keep the available capacity of the battery system at a higher level or eliminate the described damage. On the other hand, successful balancing makes it possible to use cells with relatively large differences in efficiency and / or internal resistance, which of course will also be reflected in reduced production costs.

[0017] The control device 7 is designed to perform the interventions required for balancing, which will be described in further detail below. The control device 7 may be a separate unit, or part of a higher level control system which, in addition to balancing, will provide further control functions for the battery system.

[0018] In the following, the method according to the invention will be explained for the case where the battery modules have different efficiency values, i.e., the modules of the battery string reach the lowest or highest state of charge (SoC) at different speeds. In the simplified diagram shown, the lowest state of charge is marked with 0% and the highest state of charge is marked with 100%. Additionally, for the sake of simplicity, the charging / discharging cycles shown are for the case of charging and discharging at the same constant power.

[0019] It should be mentioned that the state of charge values 0% and 100% refer to the available ranges of the respective battery modules, where, for the sake of simplicity, it is assumed that the available capacities of the battery modules connected in series are the same.

[0020] The method according to the invention relates to the cyclic operation of a battery system, where the term "cyclic operation" is defined very broadly. This is understood to mean the alternating operation of a discharging phase and a charging phase. Here, each of the two phases mentioned forms a half-cycle. In each half-cycle, the battery system can be in a standby state for any number and any length of time periods. The standby state is understood to be a state where the current flowing through the battery string is zero. Thus, a half-cycle starts with the reversal of the current direction flowing through the battery string and ends with the subsequent reversal of the current direction. A half-cycle does not necessarily have to extend over the entire state of charge range (i.e., from 0% to 100%), but can start at any first state of charge value SoC1 and end at any other state of charge value SoC2. Obviously, for a charging half-cycle, SoC1 < SoC2, and for a discharging half-cycle, SoC1 > SoC2. In fact, the half-cycles shown in the drawings of the present application usually extend over the entire state of charge range only for the sake of clarity and should not be construed as restrictive.

[0021] Figure 3 Two charging / discharging cycles of two series-connected battery modules with different efficiency values are shown. For the sake of clarity, the efficiency difference is chosen to be very high. In an actual battery system, the efficiency difference is much smaller. In Figure 3 the SoC curve of the battery module with higher efficiency is shown as a solid line, and the SoC curve of the battery module with lower efficiency is shown as a dashed line.

[0022] The more efficient module reaches a SoC value of 100% when the less efficient module is not yet fully charged. Since the current flowing through both modules is the same, the current must now be reversed and the charging half-cycle ended, so that the discharge has already begun when the less efficient module is not yet fully charged. Due to this unequal discharge starting point and the lower efficiency, the less efficient module reaches a SoC value of 0% when the more efficient module is not yet fully discharged. Since the described effects are cumulative, the SoC curves of the two modules diverge further and further as the number of cycles increases and the available capacity of the battery system continues to decrease.

[0023] The negative effects described can be avoided by taking balancing measures. Various types of possible balancing interventions are described in more detail below.

[0024] Figure 4 Two diagrams are shown, each of which is a charging cycle of two battery modules connected in series with different efficiencies. In each case, a corresponding balancing intervention is performed in the corresponding half-cycle. The balancing intervention is of the "disconnection coupling" type. With this type of balancing intervention, one or more battery modules are disconnected from the series connection for a certain period of time, so that the battery module no longer participates in the charging or discharging of other battery modules during this period. Ideally, the SoC curve of the module is horizontal when disconnected (self-discharge can be ignored).

[0025] The more efficient battery modules are disconnected during the charging half cycle, while the less efficient battery modules are disconnected during the discharging half cycle. The difference between the two diagrams is the timing of the disconnection.

[0026] In the left diagram, the relevant battery module is only decoupled when the two curves are equal to each other. For example, when charging, the battery module with high efficiency is decoupled until the battery module with low efficiency reaches the same state of charge as the battery module with high efficiency actually exhibits. The associated balancing intervention can be classified as an "equalization" type.

[0027] In the right diagram, the relevant battery module is disconnected for a longer period of time, so that the two curves of the battery modules intersect. For example, during the charging half-cycle, the battery module with higher efficiency is disconnected until the battery module with lower efficiency reaches a sufficiently high "advance" during charging, so that the battery module with higher efficiency only catches up with the battery module with lower efficiency at a SoC value of 100%. The associated balancing intervention can be classified as "overcompensation".

[0028] It should be noted that the measure of overcompensation depends of course on the SoC2 value at the end of the relevant half-cycle. As mentioned above, SoC2 does not have to be 0% or 100%, but can be anything in between. For example, if the charging half-cycle has ended at SoC2 = 80%, the more efficient battery module will have to be charged at a higher rate than Figure 4 The example shown is decoupled in a shorter time.

[0029] If a "decoupling" type of balancing intervention is to be used, it is best to design the battery string so that at most one battery module can be decoupled at a time at full charge or discharge power. If the required charge or discharge power is lower, several battery modules can be decoupled simultaneously. This reduces the time required for balancing.

[0030] To perform an "equalizing" type of balancing intervention, it is sufficient to provide the SoC values ​​of the battery modules to the battery management system that controls the intervention. For an "overcompensation" type of balancing intervention, the battery management system must perform a kind of extrapolation. For this purpose, the battery management system also requires the slopes of the SoC curves of the relevant battery modules. These slopes can be determined in a known manner from the SoC measurement values ​​at different points in time. Known smoothing filters can be applied to the slope values ​​determined in this way in order to minimize the influence of measurement errors. So-called moving average filters or such filters extended with 2nd-order exponential smoothing have proven to be effective.

[0031] Figure 5 Another balancing intervention is shown. Here, the relevant battery module is not disconnected from the series connection, but the load is connected in parallel to the relevant battery module. This causes the current flowing through the battery string to branch in such a way that only a part of the charging or discharging current flows through the battery module affected by this balancing intervention. As a result, the battery module is charged or discharged at a lower rate than if this intervention had not been taken. This means that the corresponding SoC curve is not as steep during the intervention as before or after the intervention. Figure 5 Two diagrams are shown again, of which the left diagram shows an intervention of the "equalization" type and the right diagram shows an intervention of the "overcompensation" type. The equalization intervention achieved by connecting the loads in parallel can also occasionally be referred to as the "stealing" type, since part of the charging or discharging current is used for other purposes. This approach is advantageous if the loads connected in parallel are elements that are part of the battery system and need to be supplied with energy anyway. For example, the pumps of the battery modules affected by such an intervention can be supplied with energy during the intervention by means of the diverted charging or discharging current, whereas they should have been supplied with energy from another energy source.

[0032] It should be noted that an intervention of the "overcompensation" type must be initiated in good time before the end of the corresponding half-cycle. This applies in particular to an intervention of the "stealing" type, since the battery module affected by the intervention also has a non-zero SoC slope in the case of "stealing". In principle, such an intervention of the "decoupling" type can be carried out shortly before the SoC2 value is reached, wherein the shorter the time interval before the SoC2 value is reached, the more the intervention resembles an intervention of the "equalization" type.

[0033] exist Figure 4 and Figure 5 In the simplified diagram, one balancing intervention is shown in each half-cycle. However, if necessary, more than one intervention per half-cycle can be performed. Such an intervention is usually triggered by the battery management system when the deviation between the SoC values ​​of the battery modules exceeds a predefined threshold. For example, the minimum and maximum values ​​of the measured SoC values ​​can be considered. Likewise, the variance of these values ​​can also be formed and used to trigger the balancing intervention. The smaller the threshold is chosen, the tighter the SoC curve of the battery module can be kept by increasing the execution of balancing interventions.

[0034] It should also be mentioned that Figure 5 The SoC curve shown can also be generated with the aid of interventions of the "decoupling" type. In this case, these interventions are triggered by the battery management system at a high frequency and for a correspondingly short duration. This means that Figure 5 The flat curve shown during the intervention is approximated by a large number of small steps. Here, this high-frequency sequence of decoupling and coupling steps is considered as a single equalization intervention and is classified as the type corresponding to the corresponding SoC curve (ie, "equalization" or "overcompensation").

[0035] It is also conceivable that the balancing intervention is carried out only during the charging process or only during the discharging process.

[0036] The inventors have recognized that the balancing methods known from the prior art can be improved by performing at least two balancing interventions during a half cycle, wherein the last balancing intervention performed in a half cycle is of the "overcompensation" type. This allows the SoC curves of the battery modules to coincide at the end of the half cycle. However, the inventors have also recognized that a balancing intervention of the "overcompensation" type can only be performed in such a way that it leads to the mentioned desired result if the battery management system controlling the intervention "predicts" the SoC2 value at the end of the current half cycle. The inventors have recognized that a reliable prediction of said SoC2 value can be made by a pre-trained AI (Artificial Intelligence).

[0037] The AI ​​training can be performed in different ways. One option is to perform the training on the same battery system on which the method according to the invention is also performed at a later point in time. For example, during the training phase, only balancing interventions of the "equalization" type can be performed. Alternatively, if balancing interventions of the "overcompensation" type are performed, the prediction of the required SoC2 value is performed in a different way, i.e. without the aid of AI. For example, the SoC2 value of the second to last half cycle, or the average value of the past several half cycles of the same type (i.e. discharging or charging) can simply be used.

[0038] Another option for training the AI ​​is to train it using historical data from another battery system that is as similar as possible. Another option for training is to use artificially generated data. Artificial generation can use certain known characteristics of the battery system on which the AI ​​is to be used.

[0039] The AI ​​may continue to be trained while it is used in the method according to the invention, i.e. the AI ​​may continuously "learn" during the course of the method according to the invention. Therefore, the AI ​​does not have to be perfectly trained in advance, but the training is performed until the quality of the AI's predictions is at least as good as the simple prediction options described above (using the second to last half-cycle or calculating the average from the past several half-cycles of the same type).

[0040] To make a prediction of the required SoC2 value, the AI ​​advantageously uses not only relevant variables from the relevant battery system, but also relevant variables provided by the environment, such as weather data and weather forecast data for the area over which the grid to which the battery system is connected extends. This is particularly advantageous when the battery system is used to regulate the grid.

[0041] The inventors have further recognized that it is advantageous to perform a "disconnect" type balancing intervention only when the SoC value of the battery module is in the range of 20% to 80%, i.e. in the SoC range with moderate charging or discharging currents. Thus, the switching operations required for the intervention can be performed with less wear.

[0042] Figure 6Two diagrams are shown of the SoC curves of two battery modules in which an equalizing intervention has been performed according to the method of the invention. In this case, in the left diagram, an intervention of the "disconnection coupling" type is used. In the right diagram, the intervention shown can be of the "stealing" type or caused by a high-frequency "disconnection coupling" (see above). In both diagrams, the SoC range from 20% to 80% is marked by two additional horizontal dashed lines. All equalizing interventions are carried out within the SoC range between these lines, i.e. the interventions are terminated in this range respectively (this is important for the interventions in the right diagram, where the SoC value of the relevant battery module changes during the intervention). It should be noted that in the right diagram, this criterion is only favorable if the curve shown is caused by a high-frequency disconnection coupling. In the left diagram, three interventions are carried out per half cycle, while in the right diagram, only two interventions are carried out per half cycle. Here, the last intervention of each half cycle is of the "overcompensation" type, while the other interventions are of the "equalization" type. Here, the value at the end of the SoC2 of each half cycle is 0% or 100%, respectively. However, as mentioned above, any other SoC2 value can complete the half cycle.

[0043] It should be noted that the "decoupling" type and the "stealing" type can also appear in mixed form within a half-cycle.

[0044] Finally, it should be noted that the battery management system controlling the interventions is a computer program running in the control device 7. This means that the control device 7 controls the balancing interventions and detects the SoC measured values ​​of the individual battery modules 1 for this purpose. However, other measured variables (see above) can also be used as control variables (e.g., measured values ​​of the terminal voltage of the battery module 1), since these measured values ​​are proportional to their SoC values. Here, the AI ​​for predicting the SoC2 value is part of the computer program.

[0045] List of Reference Numerals

[0046] 1 Battery Module

[0047] 2 Monomer device

[0048] 3 Storage tank equipment

[0049] 4 Measurement equipment for determining OCV

[0050] 5 Measuring equipment for determining terminal voltage

[0051] 6Bidirectional Conversion System (PCS)

[0052] 7. Control Equipment

Claims

1. A method for operating a redox flow battery system, the redox flow battery system comprising at least two battery modules (1), a bidirectional conversion system (6) and a control device (7), wherein: The battery modules (1) are connected in series and connected to the conversion system (6), and wherein each battery module (1) comprises a cell device (2) having a plurality of redox flow cells and a tank device (3) for storing electrolyte and supplying electrolyte to the cell device (2), and wherein the redox flow battery system comprises a measuring device (4, 5) for providing a measured variable, wherein the measured variable represents a measured value of the state of charge (SoC) of each battery module (1), and wherein the method comprises the following steps: - Cyclic operation of the redox flow battery system; - detecting a measured value by using the measuring device (4, 5); - performing at least two balancing interventions on the battery module (1) during a half cycle; in, The last equalization intervention performed in said half-cycle is of the “overcompensation” type and the SoC2 value used for this equalization intervention is predicted with the help of a pre-trained AI.

2. The method according to claim 1, wherein: At least one balancing intervention performed is of the "decoupling" type and is only carried out when the state of charge of the relevant battery module (1) is between 20% and 80%.

3. The method according to claim 1 or 2, wherein: At least one equilibrium intervention performed was of the "stealing" type.

4. A redox flow battery system configured to perform the method according to any one of claims 1 to 3.

5. A computer program for executing the steps of the method according to any one of claims 1 to 3.

6. A data carrier having stored thereon a computer program according to claim 5.

Images