Method and device for generating user-specific charging strategies for device batteries of technical devices

Abstract

The invention relates to a method, in particular an at least partially computer-implemented method, for providing a lower and / or upper charge state limit (uLG, oLG) for carrying out charging operations of a device battery (12), wherein the lower and / or the upper charge state limit (uLG, oLG) indicates a limit of the actual charge state of the device battery (12) at which the discharging of the device battery (12) is stopped or at which the charging is stopped; comprising the following steps: detecting (S1) operating variable curves of a device battery (12); ascertaining (S2) a usage pattern from the operating variable curves of a specified, in particular immediately preceding time period, wherein the usage pattern indicates operating features which represent a load on the device battery (12), wherein an artificial curve of operating variables can be ascertained depending on the usage pattern; determining (S3, S4, S5) the lower and / or upper charge state limit (uLG, oLG) by optimising or maximising a remaining service life of the device battery (12) with the aid of an optimisation method in that an artificial curve of operating variables, depending on the usage pattern and depending on the lower and / or upper charge state limit (uLG, oLG), is predicted for lower and / or upper charge state limits (uLG, oLG) which vary in accordance with the optimisation method; a predicted calendar-based time profile of an ageing state is ascertained with the aid of a time series-based ageing model on the basis of the artificial curve of operating variables; the remaining service life is determined on the basis of the predicted time profile of the ageing state; providing (S6) the lower and / or upper charge state limit (uLG, oLG) for the optimised or maximised remaining service life for operating the device battery (12).

Description

[0001] Description

[0002] title

[0003] Method and device for generating user-specific charging strategies for device batteries of technical devices

[0004] Technical area

[0005] The invention relates to device batteries for operating technical devices, and in particular to the use of charging strategies for charging such device batteries.

[0006] Technical background

[0007] The use of battery-operated technical devices depends significantly on the application and behavior of the device's user. Accordingly, the sequence of charging phases, discharging phases, and rest phases of the device battery varies from device to device. Furthermore, the number of charging processes is significantly determined by the load during the discharging phases. The charging processes, in particular their duration, charging cycles, initial charge states, and final charge states, are therefore significantly dependent on the energy consumption during the discharging phases, which are determined by the type of use.

[0008] Typically, the upper and lower charge limits, i.e. the maximum and minimum real charge levels, are defined for the use of a device battery, and the available real charge level range in between is mapped to a range between 0 and 100% as the displayed charge level. This means that the user is shown a fully discharged device battery with a charge level of 0% and a fully charged battery with a charge level of 100%, with the real charge level of the discharged device battery at a higher charge level (lower charge level limit), e.g., approximately 20% SOC, and the real charge level of the fully charged device battery at a lower charge level (upper charge level limit), e.g., 80% SOC. This enables battery-friendly charging orA more battery-friendly operation of the device battery can be achieved, and the additional capacity of the device battery can be released between 0% actual charge level and the lower charge level limit and between the upper charge level limit and 100% actual charge level if necessary, at the expense of increased battery aging.

[0009] Disclosure of the invention

[0010] According to the invention, a method for generating a user-specific charging strategy for device batteries of a battery-operated technical device according to claim 1 and a corresponding device according to the independent claim are provided.

[0011] Further embodiments are specified in the dependent claims.

[0012] According to a first aspect, a method, in particular an at least partially computer-implemented method, is provided for providing a lower and / or upper charge limit for carrying out charging processes of a device battery, wherein the lower and / or upper charge limit indicate a limit of the actual charge level of the device battery at which the discharging of the device battery is stopped or at which the charging is stopped; comprising the following steps:

[0013] Recording operating parameters of a device battery;

[0014] Determining a usage pattern from the operating variable curves of a given, in particular immediately past, period, wherein the usage pattern indicates operating characteristics that represent a load on the device battery, wherein an artificial curve of operating variables can be determined depending on the usage pattern;

[0015] Determining the lower and / or upper state of charge limit by optimising or maximising a remaining service life of the device battery using an optimisation method by o predicting an artificial course of operating variables depending on the usage pattern and depending on the lower and / or upper state of charge limit for lower and / or upper state of charge limits varying according to the optimisation method; o determining a predicted calendar-based temporal course of an aging state using a time-series-based aging model depending on the artificial course of operating variables; o determining the remaining service life depending on the predicted temporal course of the aging state;

[0016] Providing the lower and / or upper state of charge limit for the optimized or maximized remaining service life for operating the device battery.

[0017] Portable batteries degrade over their service life and depending on their load and use. This so-called aging leads to a continuously decreasing maximum power or storage capacity. The state of aging corresponds to a measure of the aging of energy storage devices. According to convention, a new portable battery can have an aging state (in terms of its capacity, SOH-C) of 100%, which decreases noticeably over the course of its service life. The measure of the aging of a portable battery (change in the aging state over time) depends on the individual load on the portable battery, i.e., in the case of vehicle batteries, the driver's usage behavior, external ambient conditions, and the vehicle battery type.

[0018] Operating a device battery, i.e. charging and discharging it, at real charge levels below the lower charge level limit and above the upper charge level limit, leads to increased degradation of the device battery. However, normal operation of a technical device usually does not require the device battery to be used to its full capacity. Therefore, by adjusting the lower and upper charge levels, which correspond to a real minimum permissible charge level and a real maximum permissible charge level, respectively, aging can be reduced and the service life of the device battery extended. The real charge level of a device battery indicates the ratio of the stored usable charge to the maximum storable usable charge.

[0019] It may be provided that the lower and / or upper charge level limit is displayed to a user as a displayed charge level of 0% or a displayed charge level of 100%.

[0020] A user often has a specific charging behavior, whereby a charging process is started when the average charge level falls below a certain displayed state of charge and is terminated when a certain displayed state of charge, e.g., 100%, is reached. Thus, by specifying the lower and upper state of charge limits, the actual state of charge limits between which the charging process is carried out can be influenced.

[0021] Adapting a charging strategy by setting a lower and / or upper state of charge limit is conventionally performed to provide additional battery capacity under special conditions.

[0022] Due to the individual use of the device battery, the charging behavior differs, in particular because when a lower limit of a displayed charge level is reached, the search for a charging option is triggered.

[0023] Knowing the charging behavior—that is, the level of charge at which a user attempts to initiate a charging process and the level of charge at which the user aborts or terminates the charging process—significantly determines the degradation of the device's battery. As long as the number of charging processes does not change significantly, i.e., there is no noticeable loss of convenience for the user, e.g., more than one or two additional charging processes per week, the load on the device's battery during a charging process can be influenced by specifying lower and / or upper charge limits.

[0024] The available charge range can be displayed to the user as a state of charge between 0 and 100% SOG, while the lower and upper SOG limits can correspond to real-world states of charge, for example, between 5 and 30% (e.g., 20%), or between 70 and 95% (e.g., 80%) SOG. Adjusting these lower and / or upper SOG limits thus determines the amount of charge flow during a charging process, particularly in degradation-intensive areas of very low and very high SOG, and thus the degradation of the device battery over a longer period of time.

[0025] The current aging state can be determined using conventional and known models for determining the aging state. In particular, an electrochemical aging state model can be used, which is fundamentally based on an electrochemical battery model. Such an electrochemical battery model can comprise a system of differential equations that, based on differential equations parameterized via model parameters, models internal battery states, in particular equilibrium states and, if applicable, kinetic states, using a time integration method and provides a relationship between operating parameter profiles of the device battery, namely a battery current, a battery voltage, a battery temperature, and a state of charge of the device battery, and the internal battery state. Such electrochemical battery models are known, for example, from the publications US 2016 / 023,566, US 2016 / 023,567, and US 2020 / 150,185.An aging state can be derived from the internal battery states.

[0026] According to the above method, with a suitable prediction of usage behavior, it is intended to predict the aging state of the device battery. Thus, the remaining service life of the device battery can be determined by the predicted aging state reaching a predetermined aging state limit, such as SOH-C = 70%, 75%, or 80%.

[0027] The basis for predicting the aging state is a usage pattern that reflects past usage behavior. The usage pattern indicates the type of use that leads to a certain load on the device battery and can be specified in a variety of ways. The usage pattern can make it possible to artificially specify or simulate a continuous course of load variables, such as battery current and battery temperature, and possibly battery voltage and state of charge, so that these represent a load on the device battery in a future period that corresponds to the load in the past, which is specified by the usage pattern, in particular during a specified past period of, for example, 6 months.

[0028] Furthermore, the usage pattern can define charging cycles, discharging cycles and rest cycles, each of which represents a time period of an operating variable.

[0029] The usage pattern can, for example, be determined from historical usage-dependent operating parameters of the device battery and can consist of a sequence of charging cycles, discharging cycles, and rest cycles. The cycles are determined by their average durations, power throughputs, and one or more other usage characteristics. The charging cycles can be defined by charging profiles, and the discharging cycles by discharging profiles.

[0030] The charging cycles are modeled using one or more predetermined charging curves, whereby a starting charging state and a final charging state are assumed according to a frequency distribution of the starting charging states and the final charging states of the past use.

[0031] When predicting the aging state, the charging cycles with charging processes corresponding to the lower and upper state of charge limits of the actual state of charge are then taken into account. Since the charging processes cause degradation of the device battery, which depends on the lower and upper state of charge limits of the actual state of charge, the aging of the device battery and thus its remaining service life can be influenced by appropriately varying the lower and / or upper state of charge limits.

[0032] Furthermore, the lower and / or upper state of charge limit can be taken into account as a constraint during optimization, so that the frequency of charging processes is not increased by more than a specified number compared to the frequency of charging processes in the past.

[0033] Depending on the type of use during discharge cycles, changing the upper and lower charge level limits can lead to variations in the number of charging processes. Users, based on their own usage behavior, are highly likely to initiate a charging process when the battery level falls below a certain displayed charge level. For example, the charging process may start at a displayed charge level of 30%, while the actual charge level may be 20% or 40%. The final value of the displayed charge level is also specified by the user, e.g., through charging settings, or may correspond to a displayed charge level of 100%, which corresponds to a supposedly fully charged device battery, although the actual charge level is likely to be lower than the displayed charge level.

[0034] Based on the usage pattern, which is specified in such a way that a course of the aging state can be predicted using a suitable aging state model that indicates a comparable load on the device battery, the influence of the predicted charging cycles on the degradation of the device battery can now be modeled by iteratively varying the upper and / or lower state of charge limit.

[0035] Methods for predicting artificial operating variables for portable batteries are known, for example, from DE 102022 202 882 A1 and DE 102021 212 689 A1.

[0036] Predicted aging state trajectories are obtained, from which a remaining service life can be determined, namely the time until a predetermined aging state threshold value of, for example, 80% SOH is reached.

[0037] By carrying out an optimization procedure in which a remaining service life is determined for each changed lower and / or upper state of charge limits, a remaining service life can be increased or maximized by specifying an optimized charging operation strategy for the individual user of the technical device by specifying the lower and / or upper state of charge limit at which the remaining service life is maximized.

[0038] Due to the required computing capacity, the method can be executed in a central unit remote from the device. For this purpose, the operating parameters of the technical device, in particular the device battery, such as battery current, battery voltage, battery temperature, and charge level, are transmitted to the remote central unit with high temporal resolution, and a usage pattern is created from this that reflects the resulting load on the device battery over a previous period of, for example, 6 months.

[0039] The usage pattern can then be used to appropriately predict charging cycles, discharging cycles, and rest cycles, e.g., in the form of operating parameter curves. In particular, the operating parameter curves are predicted by a sequence of characteristic charging cycles, discharging cycles, and rest cycles resulting from the previously determined usage pattern.

[0040] In general, the usage pattern can define one or more charging cycles, one or more discharging cycles and one or more rest cycles, each of which represents a time period of a given operating variable.

[0041] Furthermore, in order to determine an artificial operating variable profile, the time segments of the operating variable profiles of the one or more charging cycles, the one or more discharging cycles and the one or more rest cycles can be arranged in a row, the row being arranged in a sequence which corresponds to a sequence of cycles in the predetermined previous period or is determined therefrom, in particular on the basis of frequency considerations, the time segments of the operating variable profiles for the one or more charging cycles and the one or more discharging cycles being compressed, clipped or stretched in time in such a way that a real state of charge value results for the end times of a discharging cycle and a charging cycle, which real state of charge value is obtained from the lower state of charge limit and the average displayed starting state of charge at the start of a charging process orfrom the upper charge level limit and the average final charge level displayed at the end of a charging process.

[0042] For example, the charging cycles, discharging cycles, and rest cycles can be arranged in sequence according to a hidden Markov model predetermined by the usage pattern, wherein a frequency distribution of the starting state of charge and / or the final state of charge of charging processes is determined from the operating variable profiles of the predetermined, in particular immediately previous, period. A probability distribution for the actual starting state of charge and the actual final state of charge is determined from the frequency distribution of the starting state of charge and / or the final state of charge and depending on the lower and / or upper state of charge limit, wherein the transitions between discharging cycles (and subsequent rest cycles, if applicable) and charging cycles occur at states of charge that are randomly selected according to the probability distribution for the actual starting state of charge, and / or wherein the transitions between charging cycles (and subsequent rest cycles, if applicable)subsequent rest cycles) and discharge cycles at charge states that are randomly selected according to the probability distribution for the real final charge state.

[0043] Relevant variables are necessary to characterize the charging cycles in the usage pattern, in particular the frequency distribution of the initial state of charge and the final state of charge (the displayed state of charge). Therefore, when predicting the aging process, charging cycles are assumed to be probable based on the frequency distribution of the initial state of charge and the final state of charge. Varying the upper and lower state of charge limits (the actual state of charge) shifts the duration and charging range (charging range and charging range size) of a charging cycle, since the state of charge range between the upper and lower state of charge limits is mapped to a displayed state of charge range between 0 and 100%, and user behavior is influenced by the information or display of the displayed state of charge.

[0044] Once the upper and / or lower charge level limits have been determined according to the optimization procedure above, they can be transmitted back to the technical device to be taken into account for future charging processes and the display of the displayed charge level.

[0045] Furthermore, the upper and lower state of charge limits are used to signal the available state of charge range to the driver as a range between 0 and 100% of the state of charge. This means that the state of charge range between the lower and upper state of charge limits is mapped to a state of charge range between 0 and 100%. Brief description of the drawings

[0046] Embodiments are explained in more detail below with reference to the attached drawings. They show:

[0047] Figure 1 is a schematic representation of a vehicle with a

[0048] Vehicle battery that is in communication with a central unit;

[0049] Figure 2 is a flowchart illustrating a method for optimising the charging strategy of a vehicle depending on the historical use of the vehicle; and

[0050] Figure 3 is a diagram illustrating the generation of artificial operating variables depending on a probability distribution of a starting state of charge.

[0051] Description of embodiments

[0052] Figure 1 schematically shows a vehicle 1 as an example of a technical device. The vehicle 1 is powered by an electric motor 11, which is supplied with energy via a vehicle battery 12 as a device battery. The vehicle has a control unit 13. A battery management system 14 is provided for monitoring the vehicle battery 12 and for measuring operating variables.

[0053] The control unit 13 or the battery management system 14 uses suitable sensors to record operating variables of the vehicle battery 12, in particular the terminal voltage, the battery current, and the battery temperature, and determines the state of charge by integrating charge inflows and outflows. A communication unit 15 can transmit these operating variables as operating variable profiles to a central unit 2 remote from the vehicle.

[0054] The central unit 2 comprises a data processing unit 21 in which part of the method described below can be carried out, and a database 22 for storing data points, model parameters, states and the like.

[0055] Figure 2 shows a flowchart illustrating an exemplary procedure for determining an upper and / or lower state of charge limit. The method can be performed in conjunction with control unit 13 and central processing unit 2.

[0056] In step S1, operating variables, i.e. the battery voltage, the battery current, the battery temperature and the state of charge at cell level and / or at module level, are recorded in the vehicle battery 12 and transmitted to the central unit as time series.

[0057] Based on the operating parameter profiles, an aging state model can be parameterized and / or created in the central unit 2, which makes it possible to provide precise information about the aging state. Such a model can be further improved by using a large number of operating parameter profiles from multiple vehicles with similar vehicle batteries.

[0058] The current aging state can be determined from the operating variables using conventional and known models for determining the aging state. In particular, an electrochemical aging state model can be used, which is fundamentally based on an electrochemical battery model. Such an electrochemical battery model can comprise a system of differential equations that, based on differential equations parameterized via model parameters, models internal battery states, in particular equilibrium states and, if applicable, kinetic states, using a time integration method and provides a relationship between the operating variables of the device battery, namely a battery current, a battery voltage, a battery temperature, and a state of charge of the device battery, and the internal battery state.Such electrochemical battery models are known, for example, from the publications US 2016 / 023,566, US 2016 / 023,567, and US 2020 / 150,185. An aging state can be derived from the internal battery states. In step S2, a usage pattern can be determined in the central unit 2 from the historical operating variables, based on the type of use within a predetermined, particularly recent, period of time, e.g., 6 months.

[0059] The usage pattern can indicate the type of use in a variety of ways. For example, for discharge cycles, the energy throughput, such as the ampere-hour throughput, the duration and variation of the power drawn at respective battery temperatures and states of charge, and other characteristics can be used to describe usage. A rest cycle can be described by its average duration, average battery temperature, and average state of charge. A charging cycle can be specified by a charging curve (the course of the average charging current depending on the state of charge), a lower limit of the displayed state of charge, and an upper limit of the displayed state of charge.

[0060] For example, a hidden Markov model can be created from the sequence of discharge cycles, charge cycles and rest cycles, which allows an artificial generation of a future use, so that synthetic or artificial, ie not real, operating variable profiles can be generated, which lead to a comparable load on the vehicle battery 12.

[0061] In step S3, the usage pattern enables the generation of artificial operating parameter profiles. The usage pattern can, for example, select a time period, such as a week, a day, a month, or similar, within the predetermined, particularly immediately past, period within which the vehicle battery 12 was used on average, i.e., the time period in which the change in aging state is closest to the average of all aging state changes in the time periods. A sequence of cycles, including their durations and loads, can now be determined, e.g., in the form of an ampere-hour throughput.Each charging cycle can be assigned an operating variable profile resulting from a charging curve (profile of the maximum charging current versus the state of charge), and each discharging cycle can be assigned a varying operating variable profile corresponding to a load on the vehicle battery 12, which indicates the average load of the discharge cycles from the predetermined period. To generate the predicted artificial operating variable profiles, the operating variable profiles can be arranged one or more times in a sequence within the predetermined period. Multiple discharge cycles can also be defined and assigned operating variable profiles that have different durations and ampere-hour throughputs, so that city trips, motorway trips, and cross-country trips with significantly different durations within the predetermined period can also be taken into account in the artificial operating variable profile.

[0062] In order to take into account the upper state of charge limit varied in the optimization process described below, the operating variable curves of the charging cycles can be shortened accordingly or extended according to the charging curve in order to cause a charging stop in the artificial operating variable curve at a real state of charge, which results from the average displayed final state of charge and the upper state of charge limit and corresponds to the real state of charge at which the user on average aborts the charging process.

[0063] In order to take into account the lower state of charge limit varied in the optimization procedure described below, the operating variable curves of the discharge cycles can be shortened accordingly by temporal compression or stretched by temporal extension in order to cause a discharge stop in the artificial operating variable curve at a real state of charge, which results from the average displayed starting state of charge and the lower state of charge limit and corresponds to the real state of charge at which the user on average starts the charging process.

[0064] Alternatively, using a hidden Markov model, sequences of characteristic charging, discharging, and resting cycles (preferably multiple charging, multiple discharging, and multiple resting cycles) can be determined for random selection according to the transition probabilities between charging, discharging, and resting cycles. Operating variable profiles with a load on the vehicle battery 12 determined by the usage pattern are assigned to each of these cycles. Each cycle is then assigned a temporal segment of a specific operating variable profile, which are then arranged in sequence. As described above, the actual state of charge at which a charging cycle ends or a discharging cycle begins can be specified by temporally compressing, stretching, or shortening the temporal segments of the operating variable profiles.

[0065] The usage pattern can be used to predict operating metric trends based on usage behavior during the predetermined past period. The predicted operating metric trends can then be used to determine an aging trajectory according to the provided aging model.

[0066] When generating the predicted operating variables, lower and upper state of charge limits (uLG, oLG) are taken into account for the charging cycles. The operating variables of the individual charging cycles can be truncated at the limits, e.g., by eliminating a charging step at the edges in a stepped charging profile. Alternatively, the temporal sections of the operating variables can be compressed, truncated, or stretched to achieve the actual states of charge determined by the lower and upper state of charge limits for the start of a charging cycle or the end of a charging cycle, assuming the corresponding artificial operating variables.

[0067] The real starting charge state SLstart is then given as

[0068] SLstart = uLG + SLstart_ang / (oLG - uLG)

[0069] With SLstart_ang the average displayed starting charge level determined over several charging processes, which is shown to the user.

[0070] The real final state of charge SLend is then given as

[0071] SLend = oLG - (100 % - SLend_ang) / (oLG - uLG) with SLend_ang the average displayed final charge level determined over several charging processes, which is displayed to the user, ie often 100 % displayed charge level.

[0072] In addition to the charging curve, the usage pattern also specifies an upper and lower state of charge limit for the charging cycles. The upper and lower state of charge limits indicate a real SOG value of, for example, 20% for a displayed state of charge of 0%, and an upper state of charge limit of, for example, 80% of the real state of charge for a displayed state of charge of 100%.

[0073] In step S4, a progression of the predicted aging state is determined based on the calendar age of the vehicle battery 12, which is determined using the artificial operating variables and the aging state model. The remaining service life is determined by the point in time at which the aging state falls below a predefined aging state threshold (SOH-C).

[0074] Based on the usage pattern, the upper and / or lower state of charge limits can now be varied so that predicted operating parameter curves are based on the respective specified state of charge limits. This takes into account that, according to a frequency distribution, a user starts a charging process at a lower displayed state of charge and ends it at an upper displayed state of charge, which often corresponds to a full charge. When considering the charging cycles to determine the predicted aging state, the starting state of charge and the ending state of charge can be stochastically selected according to a probability distribution that corresponds to the frequency distribution from the usage pattern. By varying the lower and upper state of charge limits of the actual state of charge, the frequency distribution of the starting state of charge and the ending state of charge is shifted with respect to the actual state of charge.

[0075] Figure 3 schematically shows the transition of the operating variable profiles of the battery current I and the real state of charge SOC from a discharge cycle EZ1 to a charging cycle LZ and from the charging cycle LZ to a further discharge cycle EZ2. One can see the starting state of charge SOC* at the beginning of the charging cycle, which results from a random selection of the probability distribution of the starting states of charge when compiling the artificial operating variable profile and can therefore vary for further such transitions within the predicted operating variable profile. One can also see the final state of charge SOC** at the beginning of the discharge cycle, which results from a random selection of the probability distribution of the final states of charge when compiling the artificial operating variable profile and can therefore vary for further such transitions within the predicted operating variable profile.The actual charge levels of the starting charge level and the final charge level depend on the lower and upper charge level limits.

[0076] When generating predicted operating variables, operating variables for discharge cycles following a charge cycle can be configured such that the displayed state of charge at the time of the transition from the discharge cycle to the charge cycle corresponds to the specified frequency distribution. By lowering the lower state of charge limit, the transition between the discharge cycle and the charge cycle occurs at lower average real states of charge than is the case with higher lower state of charge limits, and vice versa.

[0077] By iteratively adjusting the upper and / or lower state of charge limit and subsequent prediction of the artificial operating variable curves as in step S3 and determination of the aging state curve as in step S4, the remaining service life of the vehicle battery 12 can be optimized in step S5.

[0078] When generating the predicted operating size curves, it can also be considered that the frequency of charging cycles is not significantly increased in order to avoid deteriorating user comfort. Thus, it can generally be acceptable if the frequency of charging cycles does not exceed a certain number of charging cycles per unit of time compared to the previous frequency, such as one additional charging cycle per week.

[0079] The upper and lower state of charge limits determined by the optimization can now be transmitted back to the vehicle in a subsequent step S6 and implemented there. In particular, the upper and lower state of charge limits (as real state of charge limits) determine the displayed state of charge of 100% and 0%, respectively, since this corresponds to the usable capacity range of the vehicle battery 12.

[0080] The vehicle may further be provided with a feature that allows the driver to selectively deactivate the automated charging strategy in order to increase the available battery capacity and extend the vehicle's range. In a further embodiment, the charging strategy can also be coupled with household energy consumption, for example, by taking into account excess energy from a renewable energy system used to temporarily charge the vehicle battery, or when energy from the vehicle battery 12 is made available to supply energy to various consumers in the household. In this case, the automated charging strategy can be deactivated in order to utilize a higher capacity of the vehicle battery.

Claims

Claims 1 . A method, in particular an at least partially computer-implemented method, for providing a lower and / or upper charge limit (uLG, oLG) for carrying out charging processes of a device battery (12), wherein the lower and / or upper charge limit (uLG, oLG) indicate a limit of the actual charge level of the device battery (12) at which the discharging of the device battery (12) is stopped or at which the charging is stopped; comprising the following steps: Recording (S1) operating parameter curves of a device battery (12); Determining (S2) a usage pattern from the operating variable curves of a predetermined, in particular immediately past, period, wherein the usage pattern indicates operating characteristics that represent a load on the device battery (12), wherein an artificial curve of operating variables can be determined depending on the usage pattern; Determining (S3, S4, S5) the lower and / or upper state of charge limit (uLG, oLG) by optimizing or maximizing a remaining service life of the device battery (12) using an optimization method, in which o an artificial course of operating variables is predicted depending on the usage pattern and depending on the lower and / or upper state of charge limit (uLG, oLG) for lower and / or upper state of charge limits (uLG, oLG) varying according to the optimization method; o a predicted calendar-based temporal course of an aging state is determined using a time-series-based aging model depending on the artificial course of operating variables; o the remaining service life is determined depending on the predicted temporal course of the aging state; Providing (S6) the lower and / or upper state of charge limit (uLG, oLG) for the optimized or maximized remaining service life for operating the device battery (12).

2. Method according to claim 1 , wherein a user is provided with the lower and / or the upper charge level limit (uLG, oLG) is displayed as a charge level of 0% or a charge level of 100%.

3. The method according to claim 1 or 2, wherein the usage pattern defines at which displayed starting charge level a charging process is started on average and at which displayed final charge level a charging process is ended on average.

4. The method according to any one of claims 1 to 3, wherein the aging state model comprises an electrochemical battery model comprising a time-variant differential equation system.

5. The method according to any one of claims 1 to 4, wherein the usage pattern defines one or more charging cycles, one or more discharging cycles and one or more rest cycles, each representing a time segment of a predetermined operating variable profile.

6. The method according to claim 5, wherein, to determine an artificial operating variable profile, a sequence of the time segments of the operating variable profiles of the one or more charging cycles, the one or more discharging cycles, and the one or more resting cycles is carried out, wherein the sequence is carried out according to a sequence that corresponds to a sequence of cycles in the predetermined past period or is determined therefrom, in particular based on frequency considerations, wherein the time segments of the operating variable profiles for the one or more charging cycles and the one or more discharging cycles are compressed, clipped, or stretched in time such that a real state of charge value results for the end times of a discharging cycle and a charging cycle,which results from the lower state of charge limit (uLG) and the average displayed initial state of charge (SLstart_ang) at the start of a charging process or from the upper state of charge limit (oLG) and the average displayed final state of charge (SLend_ang) at the end of a charging process.

7. The method according to claim 5, wherein a sequence of the charging cycles, discharging cycles and rest cycles is carried out according to a hidden Markov model predetermined by the usage pattern, wherein a Frequency distribution of the starting state of charge (SLstart_ang) and / or the final state of charge (SLend_ang) is determined from the operating variable curves of the specified, in particular immediately past, period, whereby a probability distribution for the real starting state of charge (SLstart) and the real final state of charge (SLend) is determined from the frequency distribution of the starting state of charge (SLstart_ang) and / or the final state of charge (SLend_ang) and depending on the lower and / or upper state of charge limit (uLG, oLG), whereby the transitions between discharge cycles and charge cycles occur at states of charge that are randomly selected according to the probability distribution for the real starting state of charge (SLstart) and / or the transitions between charge cycles and discharge cycles occur at states of charge that are randomly selected according to the probability distribution for the real final state of charge (SLend).

8. Method according to one of claims 1 to 7, wherein in the optimization of the lower and / or upper state of charge limit (uLG, oLG) it is taken into account as a secondary condition that the frequency of charging processes is not increased by more than a predetermined number within a predetermined period of time above the previous frequency of charging processes within the predetermined, in particular immediately previous, period of time.

9. The method according to any one of claims 1 to 8, wherein a selection option is provided by which the user selectively selects or deactivates the use of the optimized lower and / or upper state of charge limit (uLG, oLG).

10. Apparatus for carrying out one of the methods according to one of claims 1 to 9.

11. A computer program product comprising instructions which, when the program is executed by at least one data processing device, cause the device to carry out the steps of the method according to one of claims 1 to 9.

12. Machine-readable storage medium comprising instructions which, when Execution by at least one data processing device causes it to carry out the steps of the method according to one of claims 1 to 9.

Images