Method and apparatus for providing a predicted aging state of a device battery based on a predicted usage pattern

1 . A method for operating a vehicle including a device battery, the method comprising:

transmitting at least one operating variable of the device battery of the vehicle from a communication module of the vehicle to a central unit operably connected to the communication module, the central unit including a data processing unit;

predicting a predicted aging state of the device battery using a predetermined aging state model stored on the central unit by:

determining, using the data processing unit, successive cycles of the device battery based on the at least one operating variable of the device battery, the successive cycles including (i) at least one operating cycle of the device battery, (ii) at least one rest cycle of the device battery, and (iii) at least one charging cycle of the device battery;

determining a respective cycle profile of a plurality of cycle profiles, using the data processing unit, for each of the successive cycles so that a sequence of cycle profiles of the plurality of cycle profiles is obtained, each cycle profile of the plurality of cycle profiles characterizes a type of use of the device battery during a corresponding successive cycle and/or an amount of a load on the device battery during the corresponding successive cycle;

determining a frequency distribution of transitions from one cycle profile of the plurality of cycle profiles to a subsequent cycle profile of the plurality of cycle profiles in the sequence of cycle profiles as a hidden Markov model, using the data processing unit;

determining, using the data processing unit, a predicted sequence of cycle profiles by successively, randomly selecting cycle profiles from the sequence of cycle profiles according to the frequency distribution of transitions starting from a most recently selected cycle profile of the sequence of cycle profiles;

determining, using the data processing unit, a predicted operating variable profile by assigning at least one profile operating variable profile of a plurality of profile operating variable profiles to the cycle profiles of the predicted sequence of cycle profiles, the at least one profile operating variable profile including data representing a battery current of the device battery and/or a battery temperature of the device battery based on previously measured operating variables; and

determining the predicted aging state of the device battery by processing the predicted operating variable profile as in input to the predetermined aging state model;

determining, using the data processing unit, (i) a residual range of the vehicle based on the predicted aging state of the device battery, and (ii) an operating limit for the device battery based on the predicted aging state of the device battery, the operating limit including at least one of a current limitation and a thermal management strategy;

transmitting the residual range and the operating limit from the central unit to the communication module, and supplying the residual range and the operating limit to a control unit of the vehicle that is operably connected to the communication module and an electric motor of the vehicle; and

operating, using the control unit of the vehicle, the device battery based on the residual range and the operating limit to supply electrical energy to the electric motor of the vehicle for moving the vehicle, such that the device battery is operated according to a corresponding amount of load based on the predicted aging state.

2 . The method according to claim 1 , wherein the plurality of cycle profiles comprises (i) one or more operating cycle profiles characterizing the type of use of the device battery and/or the amount of the load on the device battery during the at least one operating cycle of the device battery, (ii) one or more rest cycle profiles characterizing the type of use of the device battery and/or the amount of the load on the device battery during the at least one rest cycle of the device battery, and (iii) one or more charging cycle profiles characterizing the type of use of the device battery and/or the amount of the load on the device battery during the at least one charging cycle.

3 . The method according to claim 2 , wherein:

determining the respective cycle profile for each of the successive cycles takes place using a rule-based classification method or a clustering method based on load features; and

the load features comprise at least one aggregated variable from the at least one operating variable.

4 . The method according to claim 3 , wherein determining the respective cycle profile for each of the successive cycles includes:

detecting the at least one rest cycle during which there is no current flow from the device battery.

5 . The method according to claim 1 , wherein the at least one operating variable of the device battery comprises the battery current, the battery temperature, a battery voltage and a charging state.

6 . The method according to claim 1 , wherein the plurality of profile operating variable profiles assigned to the cycle profiles respectively correspond to an operating variable profile of a most recent cycle assigned to a corresponding cycle profile of the plurality of cycle profiles.

7 . The method according to claim 3 , wherein the plurality of profile operating variable profiles assigned to the cycle profiles respectively correspond to an operating variable profile of the successive cycles assigned to a corresponding cycle profile of the plurality of cycle profiles and closest to centroid of an associated cluster from the clustering method.

8 . The method according to claim 1 , wherein the hidden Markov model is created, such that in which the cycle profiles of the plurality of cycle profiles form nodes that are interconnected via edges, to which are assigned frequencies of transitions from the one cycle profile to the subsequent cycle profile.

9 . The method according to claim 1 , wherein determining the predicted aging state based on the predicted operating variable profile is performed using the predetermined aging state model comprising an electrochemical model which is formed by a non-linear differential equation system and is can be solved via time integration.

10 . The method according to claim 1 , further comprising:

performing predictive maintenance on the vehicle based on the residual range and the operating limit.

11 . The method according to claim 1 , further comprising:

determining, using the data processing unit, a remaining residual service life of the device battery based on the residual range and the operating limit; and

increasing or decreasing a number of remaining rapid charging cycles of the device battery and/or blocking rapid charging cycles of the device battery based on the remaining residual service life.