Method for determining an actual state of charge of at least one battery cell of a battery device

Abstract

The present invention relates to a method for determining an estimated actual state of charge (SOC) of at least one battery cell (110) of a battery device (100), characterised by the following steps: - measuring the cell current (ZA) of the battery cell (110) over at least one time unit (TE); - determining a current actual battery charge (IBL) of the battery cell (110) based on an initial battery charge (ABL) and the detected cell current (ZA); - determining a current actual battery voltage (IBS) of the battery cell (110) by means of a non-linear, double capacitor model (NDC) based on the determined actual battery charge (IBL), - determining the actual state of charge (SOC) of the battery cell (110) based on the determined actual battery voltage (IBS).

Description

[0001] Method for determining an actual charge level of at least one battery cell of a battery device

[0002] The present invention relates to a method for determining an estimated actual charge level of at least one battery cell of a battery device, a computer program product for carrying out such a method and a determination device for carrying out such a method.

[0003] It is generally known that determining the actual charge level of a battery cell is crucial for the operation of a battery device. The actual charge level is the so-called SOC, also called state of charge. The state of charge is usually expressed as a percentage and indicates the proportion of the charge stored in the battery cell and / or battery device with reference to the maximum storable charge in this battery cell. In particular, this is specified in ampere-hours per ampere-hour or in watt-hours per watt-hour, so that the percentage charge level can be specified as the SOC and thus as the actual charge level. For example, the document CN 111060822 A is known from the prior art, which describes an SOC estimation method based on “model switching and fusion”.

[0004] This actual charge level is crucial for effectively controlling, for example, the charging strategy, operating strategy, temperature control strategies, or similar with a battery management system. The actual charge level depends on a variety of factors. A crucial factor is that an aging effect can occur in battery cells when they are in use in a vehicle or other application situations. These aging effects usually lead to a decrease in the performance and, in particular, the storage capacity of the respective battery cell over the duration of use. However, this decrease is a highly complex process, so that aging can only be actually calculated with great effort. Known methods are therefore based on rough model assumptions that use the so-called state of health (SOH for short).This health status is then estimated based on a rough model and used as the remaining residual capacity and thus the remaining maximum capacity of the respective battery cell to determine the actual charge level. As can be seen from the above explanations, a rough estimate of the health status, and in particular of the actual charge level, is subject to corresponding errors and inaccuracies. This leads to unpredictable deviations between the actual chemical operating state of the respective battery cell and the additional knowledge of the information used regarding the actual charge level and health status in the battery management system. The greater these deviations, the more serious the potential for misuse. For example, the battery management system may assume a charge level that is higher than the actual charge level of the battery cell.In such a case, this would result in any calculated and displayed remaining range no longer matching the actual remaining range. Conversely, the remaining range calculated and displayed would be too low, meaning the user would not be able or able to access the actual available charge in the battery cell at all.

[0005] The object of the present invention is to at least partially remedy the disadvantages described above. In particular, the object of the present invention is to provide, in a cost-effective and simple manner, the most accurate estimate possible of the health status and / or the actual charge level of at least one battery cell.

[0006] The above object is achieved by a method having the features of claim 1, a computer program product having the features of claim 10, and a determination device having the features of claim 11. Further features and details of the invention emerge from the subclaims, the description, and the drawings. Features and details described in connection with the method according to the invention naturally also apply in connection with the computer program product according to the invention and the determination device according to the invention, and vice versa, so that reference is or can be made to the individual aspects of the invention with regard to the disclosure. According to the invention, an estimated actual charge level of at least one battery cell of a battery device is determined. This method is characterized by the following steps:

[0007] - detecting the cell current of the battery cell over at least one time unit,

[0008] - Determining a current actual battery charge of the battery cell based on an initial battery charge and the detected cell current,

[0009] - Determining the current battery voltage of the battery cell using a non-linear double capacitor model (NDC model) based on the determined actual battery charge,

[0010] - Determining the actual charge level of the battery cell based on the determined actual battery voltage.

[0011] The core concept of the invention is based on the measurement and use of parameters that are generally commonly measured in battery devices. This includes at least the cell current of the battery cell. The direction of the current flow is irrelevant, i.e. whether the battery cell is currently being charged or discharged, i.e. whether the cell current has a positive or negative sign. The length of the time unit for a method according to the invention is also fundamentally irrelevant, but is preferably kept constant, particularly when the method is repeated over a longer period of time. The shorter the time unit selected, the higher the computational requirements, but also the more precise the resolution when carrying out a method according to the invention.The step of determining the current actual battery charge based on an initial battery charge and the detected cell current can also be referred to as an estimation of the current actual charge level based on a previous iteration for determining the actual charge level.

[0012] The next step involves determining the current actual battery charge, i.e. the current charge situation, for example in ampere hours or watt hours of the battery cell. For this purpose, a relatively simple physical relationship is used, namely the recorded cell current is added to or subtracted from an initial battery charge, depending on whether the battery cell is currently being charged or discharged. In the simplest case, a higher actual battery charge can be achieved by supplying a cell current from an initial battery charge, or conversely, a negative cell current from an initial battery charge can lead to a lower actual battery charge by discharging the battery cell. However, these are still absolute values, which in this step cannot yet reflect the actual charge level of the battery cell. A further step in the method according to the invention is provided for this purpose.

[0013] In this further step, non-linear modeling is performed. Using a non-linear, double-capacitor model, the actual battery voltage is determined based on the determined actual battery charge. The actual battery voltage now uses, in particular, the non-linear double-capacitor model, which is also abbreviated to the NDC model below. For the purposes of the present invention, the term "double-capacitor model" refers to all models with at least two combinations of resistor and capacitor. Technically, this also includes models with three resistor-capacitor combinations or even more such combinations. This NDC model models and simplifies the battery cell or even a complete battery device using a model circuit with two capacitor and resistor combinations.This NDC model is based on the fact that a voltage can be tapped at each of these two capacitor / resistor switching sections, allowing non-linear behavior of the battery device and / or the battery cell to be reproduced by this NDC model. In particular, such an NDC model uses parameters as functions of current and / or temperature and thus differs from simple resistor-capacitor models in which the parameters are independent of the current but dependent on the charge level. From the actual battery charge, an actual battery voltage that is very close to reality can now be determined, even with non-linear considerations. Based on this actual battery voltage, the actual charge level of the battery cell is then determined in the final step and preferably output.

[0014] By combining both real measurement and non-linear modeling using the NDC model in a method according to the invention, the accuracy in determining the actual charge level is significantly improved. As explained at the beginning, increased accuracy of the actual charge level, i.e., a greater approximation of the determined actual charge level to the actual chemical situation in the respective battery cell, results in improved controllability of the battery cell. This makes it possible to control the battery cell, as well as a device powered by the battery cell or battery device, more precisely. For example, achievable ranges with a vehicle or similar possible uses of the battery device can be predicted more precisely, and usability is accordingly improved.

[0015] A method according to the invention can also be referred to as a combination of a direct measuring method and an indirect measuring method. The method thus combines direct measuring steps, in which, for example, the cell current is recorded directly on a real battery cell. This involves an actual, physical measurement of the cell current. Indirectly, a series of indirect measuring steps are then carried out on the basis of this direct measurement result. These use the measured cell current as a starting point and, via the intermediate steps of the actual battery charge and the actual battery voltage, determine the actual charge level. The actual charge level output in this way now corresponds to a real value in a specific operating situation of the battery cell, which would otherwise not be directly measurable or only with increased effort.Thus, such a method can be used as part of a control device in a battery management system (BMS). The determined values ​​for the actual charge level can be made available for monitoring connected consumers, for example an electric drive motor of a vehicle. It can be advantageous if, in a method according to the invention, in addition to the cell current, the cell temperature of the battery cell is also recorded over at least one time unit and is used together with the cell current as a basis for determining the current actual battery charge. In other words, the cell temperature represents an additional input variable which can, for example, be taken into account in a physical context when determining the actual battery charge.Cell temperature is a very easily measurable physical unit, which is already available, particularly in known battery devices and / or battery cells, with a corresponding measurement capability. This now allows for further improved accuracy of the output information, and in particular of the determined actual charge level, to be ensured in a cost-effective and simple manner by expanding the input parameters.

[0016] Further advantages can also be achieved if, in a method according to the invention, the steps of detection and determination are repeated for at least one further time unit. While even a single implementation of a method according to the invention leads to an actual charge level that is as accurate as possible for that point in time, repetition can provide step-by-step or even essentially continuous monitoring. For example, the respective further time unit can follow directly on from the first time unit, so that quasi-continuous monitoring is possible using the method according to the invention. However, there can also be a process-free period between the two time units, so that, for example, the method according to the invention is expediently carried out at intervals of every 5 minutes or even more frequently in order to determine the actual charge level at the respective timed point in time.However, it is preferred to provide a process for the method according to the invention that is as continuous as possible, i.e. preferably directly repeating, with adjacent time units.

[0017] It can also be advantageous if, in a method according to the invention, when the time unit is repeated, this time unit corresponds to or essentially corresponds to the length of the first time unit. In other words, the time units of the repetitions are all the same or essentially the same length, for example, a few seconds or even just a few milliseconds. The equality of the individual time units during the repetition leads to improved comparability of the determined values, in particular the determined actual charge levels. Furthermore, the algorithmic relationships that are carried out in the respective determination step do not have to be adapted to different time units, but can be designed to be essentially identical in terms of their characteristics and also in terms of their computing time requirements.

[0018] It can also be advantageous if, in a method according to the invention, at least one of the determined and / or recorded parameters is compared with the respectively correlating, determined and / or recorded parameter from the previous implementation when repeated, in particular is checked for plausibility. This makes it possible to identify certain incorrect measurements or incorrect determinations at an early stage and, in particular, to output them as errors before a correspondingly incorrect determination of the actual charge level would be output. In other words, an error detection and / or plausibility check is now additionally integrated into the method, which further improves the accuracy of a method according to the invention and thus its usability. One possible implementation for this is to use a recursive determination of the least square integral and to subtract the integrated determined actual charge level from this.

[0019] Further advantages can be achieved if, in a method according to the invention, the determined actual charge level of the battery cell is used as an input parameter for determining an estimated state of health of the battery cell. As already explained at the beginning, in addition to information about the current actual charge level, the current state of health of the battery cell is also of key importance. In this way, for example, an aging value (dC) can be determined, which provides information about the aging rate and / or the aging state of the battery cell. Well-known algorithms such as the RLS algorithm can be used here. This is the so-called Recursive Least Squares Algorithm, which involves a recursive calculation of the least squares. Of course, other algorithms can also be used to determine the state of health.The key advantage of integration is that, based on a higher accuracy of the input parameter in the form of the actual charge level, the estimation accuracy of the health status is significantly improved. This step is preferably performed for each time unit, since this allows not only the actual charge level but also the actual health status to be determined per time unit per run of a method according to the invention.

[0020] A further advantage can be achieved if, in a method according to the invention, said method is carried out for at least two battery cells of a battery device, in particular for all or substantially all of the battery cells of a battery device. In this case, a distinction can fundamentally be made between two variants. For example, the method can be carried out in parallel for two or more battery cells, such that a specific actual charge level is determined for each battery cell that undergoes a method according to the invention. Alternatively, it is also possible for the method to be carried out jointly for two or more battery cells, i.e. for the plurality of battery cells under consideration, in particular all of the battery cells of the battery device, a common actual charge level of this number of battery cells is determined.

[0021] It can be advantageous if, in a method according to the invention, specific actual charge levels are determined for each of the at least two battery cells. As explained in the previous paragraph, this allows for very high resolution across the individual battery cells and, in particular, information about imbalances in the charge levels and aging states across the battery device. This precise monitoring brings control advantages, but leads to increased measurement effort as well as increased computational effort when implementing a method according to the invention. A combination with subsets of individual battery cells, which can be arranged, for example, in several battery modules in a battery device, is of course also possible as an interim solution.

[0022] It can also be advantageous if a method according to the invention determines a general actual charge level for the battery device. This can be done alternatively or in addition to the specific actual charge levels of individual battery cells and / or individual battery modules. This higher-level information allows for the provision of comprehensive information for the battery device with fewer measurements and, above all, less computational effort.

[0023] A further subject of the present invention is a computer program product comprising instructions which, when executed by a computer, cause the computer to perform the steps of a method according to the invention. Thus, a computer program product according to the invention provides the same advantages as those explained in detail with reference to a method according to the invention.

[0024] The present invention also relates to a determination device for determining an estimated charge level of at least one battery cell of a battery device. Such a determination device is characterized in that a detection module is provided for detecting the cell current of the battery cell over at least one time unit. A charge module is used to determine a current actual battery charge based on an initial battery charge and the detected cell current. A voltage module can be used to determine a current actual battery voltage using a non-linear, double capacitor model (NDC model) based on the determined actual battery charge. Furthermore, a charge level module is provided for determining the actual charge level of the battery cell based on the determined actual battery voltage.The detection module, the charge module, the voltage module, and / or the charge level module are designed specifically for implementing a method according to the invention. Thus, a determination device according to the invention also offers the same advantages as those explained in detail with reference to a method according to the invention.

[0025] Further advantages, features, and details of the invention will become apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. They show schematically:

[0026] Fig. 1 shows an embodiment of a method according to the invention and a determination device according to the invention,

[0027] Fig. 2 shows an embodiment of a method according to the invention over several time units,

[0028] Fig. 3 shows a further embodiment of a determination device according to the invention,

[0029] Fig. 4 shows a further embodiment of a determination device according to the invention and

[0030] Fig. 5 shows a further embodiment of a determination device according to the invention.

[0031] Figure 1 schematically shows a determination device 10 according to the invention. This device has a detection module 20, which has sensor communication with a battery cell 110 via sensors (not shown in detail). In the simplest case, at least the cell current ZA is measured and recorded using this sensor (not shown), and then made available by the detection module 20 as cell current ZA to the charging module 30. The charging module 30 is thus able to determine an actual battery charge IBL based on an existing and, for example, previously determined initial battery charge ABL and to transmit it to the voltage module 40. The voltage module 40 can access a non-linear, double capacitor model, also called an NDC model, and uses this on the basis of the actual battery charge IBL to determine an actual battery voltage IBS.This actual battery voltage IBS is now transferred to the charge level module 50, which determines an actual charge level SOC on this basis, in the simplest case by means of a lookup table or a map in which the corresponding actual charge level SOC is stored for each determined actual battery voltage IBS.

[0032] This output is significantly more accurate than with the known estimation methods, since an NDC model is used in the determination device 10 when the method according to the invention is carried out.

[0033] Figure 2 shows a schematic illustration of the time sequence in more detail. This time sequence includes a repetition over three time units TE, shown schematically and as an example. In the first time unit TE, the charging current remains essentially constant as the positive cell current ZA. The actual battery charge IBL increases from an initial battery charge ABL, with the determination taking place at the end of the first time unit TE and a corresponding sudden increase or gradual or step-by-step increase being recognizable. This occurs in the same way with a smaller increase over the two further time units TE, over which the cell current ZA is no longer constant, but slowly decreases. Of course, further repetition loops and even more complex current curves for the cell current ZA can also be taken into account here.

[0034] Figure 3 shows a determination device 10 that further develops the embodiment of Figure 1. Here, an additional sensory determination and recording of the cell temperature ZT is performed for the identical battery cell 110. This cell temperature ZT is also forwarded to the charging module 30 along with the cell current ZA and is now also taken into account when determining the actual battery charge IBL. The accuracy of the run and the overall determination of the actual charge level SOC by the charge level module 50 is thus further increased.

[0035] Figure 4 also further develops this embodiment of Figure 3, wherein several battery cells 110 of a battery device 100 are considered in parallel. Even if the determination device 10 is shown here only once with only one sensor, the determined parameters can of course also be adopted and passed on in parallel from several battery cells 110. Also shown here is an internal feedback loop, in which the determined actual battery charge IBL is now set internally in the process for the next time unit TE as the initial battery charge ABL and, here as an example, is stored in the charging module 30 for the next time unit TE.

[0036] Figure 5 shows a variant and further development of the determination device 100, wherein additional processing takes place at the end of the method. Thus, the determined actual charge level SOC is passed on as input parameter EP to a health module 60, which then uses known algorithms to determine a health level SoH and can output it for this battery cell 110 or even an entire battery device 100. Due to the increased accuracy in estimating the actual charge level SOC when using the NDC module, this increased accuracy can also be used for a more precise estimation of the health level SoH.

[0037] The above explanation of the embodiments describes the present invention exclusively by way of examples.

[0038] List of reference symbols

[0039] 10 Determination device

[0040] 20 Recording module

[0041] 30 charge module

[0042] 40 voltage module

[0043] 50 charging module

[0044] 60 Health Module

[0045] 100 battery device

[0046] 110 battery cells

[0047] ZA cell current

[0048] ZT cell temperature

[0049] TE time unit

[0050] IBL actual battery charge

[0051] ABL output battery charge

[0052] IBS actual battery voltage

[0053] SOC actual charge level

[0054] NDC non-linear, double capacitor model

[0055] EP input parameters

[0056] SoH health status

Claims

Patent claims 1 . A method for determining an estimated actual state of charge (SOC) of at least one battery cell (110) of a battery device (100), characterized by the following steps: - detecting the cell current (ZA) of the battery cell (110) over at least one time unit (TE), - determining a current actual battery charge (IBL) of the battery cell (110) based on an initial battery charge (ABL) and the detected cell current (ZA), - determining a current actual battery voltage (IBS) of the battery cell (110) by means of a non-linear, double capacitor model (NDC) based on the determined actual battery charge (IBL), - Determining the actual state of charge (SOC) of the battery cell (110) based on the determined actual battery voltage (IBS).

2. Method according to claim 1, characterized in that in addition to the cell current (ZA) the cell temperature (ZT) of the battery cell (110) is also recorded over the at least one time unit (TE) and is used as a basis for determining the current actual battery charge (IBL) together with the cell current (ZA).

3. Method according to one of the preceding claims, characterized in that the steps of detection and determination are repeated for at least one further time unit (TE).

4. Method according to claim 3, characterized in that during the repetition the time unit (TE) corresponds or substantially corresponds in length to the first time unit (TE).

5. Method according to one of claims 3 or 4, characterized in that during the repetition at least one of the determined and / or recorded parameters is compared with the respectively correlating, determined and / or recorded parameter from the previous implementation, in particular is checked for plausibility.

6. Method according to one of the preceding claims, characterized in that the determined actual charge level (SOC) of the battery cell (110) is used as an input parameter (EP) for determining an estimated state of health (SoH) of the battery cell (110).

7. Method according to one of the preceding claims, characterized in that it is carried out for at least two battery cells (110) of a battery device (100), in particular for all or substantially all battery cells (110) of a battery device (100).

8. The method according to claim 7, characterized in that specific actual charge levels (SOC) are determined for each of the at least two battery cells (110).

9. Method according to one of claims 7 or 8, characterized in that a general actual charge level (SOC) is determined for the battery device (100).

10. A computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of a method having the features of any one of claims 1 to 9.

11. Determination device (10) for determining an estimated actual charge level (SOC) of at least one battery cell (110) of a battery device (100), characterized by a detection module (20) for detecting the cell current (ZA) of the battery cell (110) over at least one time unit (TE), a charge module (30) for determining a current actual battery charge (IBL) based on an initial battery charge (ABL) and the detected cell current (ZA), a voltage module (40) for determining a current actual battery voltage (IBS) by means of a non-linear, double capacitor model (NDC) based on the determined actual battery charge (IBL), and a charge level module (50) for determining the actual charge level (SOC) of the battery cell (110) based on the determined actual battery voltage (IBS), wherein the detection module (20), the charge module (30),the voltage module (40) and / or the charge level module (50) is designed to carry out a method having the features of one of claims 1 to 9.,

Images