EP4305699A2

EP4305699A2 - Method for monitoring a battery system

Info

Publication number: EP4305699A2
Application number: EP22710551.7A
Authority: EP
Inventors: Jan Philipp Schmidt
Original assignee: Mercedes Benz Group AG
Current assignee: Mercedes Benz Group AG
Priority date: 2021-03-08
Filing date: 2022-03-02
Publication date: 2024-01-17
Also published as: CN116918133A; WO2022189221A3; KR20230128544A; JP2024507985A; WO2022189221A2; US20240113343A1; DE102021001217A1

Abstract

The invention relates to a method for monitoring a battery system comprising at least one battery module having a plurality of individual battery cells, a battery management system (BMS) and at least one sensor arrangement (ASIC), in which method the battery management system (BMS) is inactive in an idle mode of the battery system, the sensor arrangement (ASIC), in the idle mode, monitoring at least one measurement value and comparing it with predefined values, and in the event of a deviation of the measurement value from the predefined values the idle mode is ended and the battery management system (BMS) is activated by a wake-up signal. The method according to the invention is characterised in that the wake-up signal is additionally alternatively triggered by the expiration of a predefined wake-up time (ΔTwakeUp), wherein, in the event of the wake-up signal being triggered by an expiration of the predefined wake-up time (ΔT_wakeUp), a new wake-up time (ΔT_wakeUp) as well as new predefined values (U_lim, T_lim) are determined and predefined before the battery management system (BMS) reverts to the idle mode, and, in the event that the wake-up signal is triggered by a deviation of the at least one measurement value from the at least one predefined value (U_lim, T_lim), an emergency mode is started.

Description

Procedure for monitoring a battery system

The invention relates to a method for monitoring a battery system according to the type defined in more detail in the preamble of claim 1.

Battery systems with at least one battery module, which is based on a plurality of individual battery cells, are known from the prior art. Such battery systems can be used stationary. They are increasingly being used as traction batteries to power an at least partially electrically powered vehicle. In order to achieve a high power density, the individual battery cells in the battery systems are often constructed using lithium-ion technology.

On the one hand, this technology allows a high energy content of the individual battery cells, on the other hand, the high energy content means that the risk potential is relatively high in the event of a defect. One of the dangers, which in the worst case can lead to a battery fire, is the so-called thermal runaway of the battery. When the battery is active, the battery, in particular the temperature, the voltage and other variables, is therefore typically monitored via a battery management system. A thermal runaway can then be detected early enough and appropriate responses can be initiated to actively counteract serious damage or a serious hazard from the battery.

In practice, however, a significant proportion of such thermal events in batteries occur after charging and during the battery's inactive phase. The battery management system is then typically switched off or is in a sleep mode, so the thermal runaway is not detected and accordingly no countermeasures can be taken.

More recent developments, such as those known from DE 102019215 812 A1, address this problem by using a sensor arrangement or multiple sensor arrangements in addition to the battery management system. Such sensor arrangements can record measured values of the battery and at least compare them with default values with a far lower energy requirement than that of the battery management system. If critical values occur here, this is recognized accordingly and the battery management system can be woken up via a wake-up signal in order to then initiate further monitoring and, if necessary, the initiation of countermeasures. The cited publication further describes that the sensor arrangement has its own memory in order to provide the battery management system with a corresponding database after it has been woken up from sleep mode, in order to be able to assess the situation as quickly as possible.

In practice, it is the case that such sensor arrangements with their associated electronics, which are typically designed as an application-specific integrated circuit (ASIC), corresponding threshold values for generating a wake-up signal can be specified. If these threshold values are set to the generally customary limits of, for example, 2 volts per cell and/or 80° C., then the battery management system will wake up extremely late in the event that an actual thermal runaway is present. Since a certain amount of time is also required to fully "boot up" the battery management system, valuable time is lost, which in the event of a thermal runaway can mean that countermeasures initiated by the battery management system are often implemented too late to actually prevent the thermal runaway .

The object of the present invention is therefore to create an improved method for monitoring a battery system with a wake-up of the battery management system, which allows a very high level of security with acceptable energy efficiency. According to the invention, this object is achieved by a method having the features in claim 1, and here in particular in the characterizing part of claim 1. Advantageous refinements and developments result from the dependent subclaims.

The method according to the invention for monitoring a battery system with at least one battery module with a plurality of individual battery cells includes a battery management system and a sensor arrangement, which is typically embodied in the form of sensors and an application-specific integrated circuit. In the real structure, several of these sensor arrangements are provided, which respectively monitor one or some of the individual battery cells accordingly. If the battery system is not used, for example because a vehicle in which the battery system is used according to a preferred application as a traction battery is not moving, then the battery management system goes into a sleep mode to save energy, in which it typically does not require its own electrical energy. Meanwhile, the at least one sensor arrangement continues to record at least one measured value for monitoring the battery system. In particular, these measured values can include voltages and temperatures. The measured values recorded are compared with default values and, in the event of a deviation, a wake-up signal is generated for the battery management system and this is woken up or activated again. All of this is known in principle from the prior art.

In order to ensure that the wake-up takes place as early as possible so that it is still possible to intervene so that a thermal runaway of the battery system can be prevented, the invention now provides that, in addition to a wake-up signal due to a violation of a default value by the at least one measured value the wake-up signal is also triggered by the expiry of a specified wake-up time, in which case, in the event that the wake-up signal is triggered by the specified wake-up time having expired, a new wake-up time and new default values are determined and specified by the battery management system before the battery management system goes into sleep mode again changes. If, on the other hand, the wake-up signal is triggered by the detected measured value violating the default value, the battery management system is woken up and an emergency mode is started. The method according to the invention thus makes it possible to wake up the battery management system not only in an emergency, but also as a function of a wake-up time to be determined. This makes it possible to track the default values as closely as possible as threshold values for monitoring the battery system in order to ensure that the battery management system is woken up as early as possible in the event of an emergency. This can then verify the emergency in emergency mode and initiate appropriate measures in order to be able to react extraordinarily quickly and efficiently to a potential emergency, so that a thermal runaway of the battery system can still be intercepted in good time in almost all cases that occur.

According to a very advantageous development of the method according to the invention, gradients of the time profile of the at least one measured value and/or fixed changes of the at least one measured value are included in the determination of the new wake-up times and/or the new default values. In this way, predictions about the course of the measured values to be expected based on the gradients can be used to track the limits for waking up the battery management system in an emergency. Together with specified change values, it can be determined when the gradient reaches this change value. Depending on the measured value, the actual curve to be expected then runs below or above this gradient, so that normally the new default value is not to be expected to be exceeded before the new wake-up time is reached. If this nevertheless occurs, an emergency must be assumed.

This variant therefore offers the possibility of tracking the limits very closely. On the one hand, this reduces the occurrence of unwanted false alarms and, on the other hand, if the limits are violated, an emergency can be identified relatively quickly by waking up the battery management system and verifying the potential emergency. This creates time to be able to react very quickly, for example by switching off cell modules, ramping up the cooling or similar measures. In most cases, this reliably prevents the thermal runaway from spreading to the entire battery system.

According to a further very advantageous embodiment of the method according to the invention, it can also be provided that a new wake-up time and new default values are individually determined and specified for each sensor arrangement. Each of the Sensor arrays typically monitor one or a few individual battery cells.

These can react differently to events that have occurred before, such as a journey of the vehicle in which the battery system is used according to a preferred use as a traction battery. This typically has to do with the manufacturing tolerances of the individual battery cells. This particularly advantageous embodiment of the method according to the invention and the individual adjustment of wake-up times and default values can therefore be used to react to such irregularities within the battery. This ensures that some individual battery cells, which may deviate due to manufacturing tolerances, do not “accidentally” trigger an emergency alarm and wake up the battery management system, although their values in relation to these special individual cells are still considered to be quite uncritical.

According to a very advantageous development of the method according to the invention, a permitted change in the measured value is defined on the basis of the current values in order to determine the new wake-up times and the new default values. In addition, the highest and/or the lowest value detected by the sensor arrangement for an individual battery cell is determined. On this basis, a measured value limit is then calculated for the next wake-up and a gradient of the measured value is calculated at the current point in time. From these limit values in combination with the gradient, a wake-up time can then be calculated as the new wake-up time at which the gradient reaches the calculated limit of the measured value. The limit as the default value and the wake-up time as the new wake-up time are then used to "instruct" the respective sensor arrangement individually, according to the embodiment described above, in order to then put the battery management system, which has carried out these calculations, back into sleep mode .

According to a further very advantageous embodiment of this idea, the battery management system can carry out further diagnoses and calculations to validate and/or check the plausibility of the determined values, ie the new wake-up time and the new default values, in order to identify and intercept potential incorrect calculations.

According to a very advantageous further development of the method according to the invention, at least the temperatures and/or monitor cell voltages. A thermal runaway can already be determined using these values today. However, the method can be improved in the sense described above due to the close tracking according to the invention of the limits for detecting such a thermal runaway. In addition to these particularly important values of temperature and cell voltage, other values are fundamentally also conceivable, for example the pressure value described in the prior art mentioned at the outset or the like.

Depending on the measured value, the measured value limit can be calculated by adding or subtracting the specified change to the current measured value. According to a very favorable development of this idea, the sum or difference can also be calculated with an offset value, which is based, for example, on measurements within the battery system, such as the current state of charge, or is specified depending on the characteristics of the battery system. This offset value can also be calculated for the battery system via the battery management system, if this is active, and can adapt over time or to different situations, such as states of charge. This creates a quasi-self-learning offset value that is able, for example, to map and take into account aging effects in the battery system. This value can preferably be generated individually for each of the sensor arrangements in order to be able to take into account different aging effects in the individual cells of the battery system or its battery modules.

According to a very advantageous development of the method according to the invention, the emergency mode includes further measurements and/or diagnostics by the battery management system, with countermeasures being initiated and a warning message being generated in the event of a confirmed emergency, and with the Battery management system is kept awake and the at least one measured value is monitored with closely specified default values. Keeping the battery management system awake after a potential but unconfirmed emergency increases safety so that the battery system can then be actively monitored using the measurement methods available in the battery management system, such as impedance measurement or the like. In this way, it can be ruled out that the unconfirmed emergency was already the indication of an emergency, which, however, occurred at a very early stage condition is that it could not be verified by the battery management system either. By keeping the battery management system awake in such a case and closely monitoring the battery system, it is now possible to react quickly if the supposed but unconfirmed emergency turns into an actual emergency.

According to a very advantageous development of the idea, the emergency mode can be terminated after a specified period of time in which the narrow specified threshold values have not been violated, so that after a sufficient period of monitoring, the battery management system returns to the sleep mode of the battery system with continuous monitoring returns solely through the sensor arrays, and is therefore extraordinarily energy-efficient.

Further advantageous refinements of the method according to the invention also result from the exemplary embodiment, which is explained in more detail below with reference to the figures.

show:

1 shows a block diagram to clarify the sequence of the method according to the invention;

FIG. 2 shows a block diagram to explain the process of limit tracking for each sensor arrangement; FIG.

3 shows a diagram of the cell voltage over time to clarify the method using the example of monitoring the cell voltage;

4 shows a diagram of the cell temperature over time to clarify the method according to the invention using the example of the temperature during cooling of the battery system; and

FIG. 5 shows a representation analogous to that in FIG. 4 when the battery system is warmed up.

In the representation of FIG. 1, the basic course of the method according to the invention is shown in a block diagram. The actual limit monitoring, i.e. the acquisition of the measured values from a Battery system and monitoring whether these measured values violate specified default values, which are also referred to below as limits. If this is the case, the ASIC generates a wake-up signal by activating the box marked "wake up". The BMS battery management system is then activated accordingly and, if no emergency is detected and verified by the BMS battery management system, the limits are updated, which leads on the one hand to a new limit or default value and on the other hand to a new wake-up time based on this default value. The new default values are sent to the limit monitor of the respective ASICs, with the tracking of the limits being carried out individually for each ASIC sensor arrangement installed in the battery system. At the same time, the wake-up time is set accordingly so that a time-controlled wake-up signal can be sent to the BMS battery management system as an alternative and in addition to a wake-up signal from limit monitoring. The battery management system BMS then returns to sleep mode and remains there until either a violation of the closely tracked limits has been detected by the limit monitor or the next wake-up time has been reached, which leads to a renewed awakening and typically tracking of the limits in the battery management system BMS leads.

In FIG. 2, the tracking of the limits is specified in more detail in a conceivable exemplary embodiment in a further block circuit diagram. The boundary conditions from a thermal point of view, for example the cooling, the ambient temperature and the like, are taken into account and the maximum cell temperature for each individual sensor arrangement or its ASIC is determined accordingly. This temperature monitoring then leads to a determination and prediction of the temperature course to be expected using a gradient. For this purpose, value changes ΔT _th and ΔT _th.max are specified accordingly. Based on this data and the gradient, the sleep duration is calculated and a new wake-up time is set and new temperature limits are calculated, which, as already explained in the block diagram above, are then sent to the ASICs and set there accordingly. The entire sequence is preferably carried out individually for each ASIC.

A comparable sequence for the cell voltage is described in the lower half of the block circuit diagram in FIG. The basis is the determination of the minimum Cell voltage per sensor arrangement or the ASIC assigned to it. Here, too, a gradient is determined on the basis of the measured value and a parameter for the maximum change in voltage ΔU _th is specified accordingly. On the basis of this gradient and this default value, a sleep duration is then calculated and a wake-up time is set, and a new voltage limit is calculated as the default value and sent to the respective ASIC, where it is used accordingly for future limit monitoring.

The diagrams in FIGS. 3 to 5 now show the voltage over time and temperatures over time in various exemplary cases to explain the basic process sequence. The measured values are shown with a dash-dotted line, the gradients in the respective points are shown with the thin solid lines and a potential deviation of the respective measured value in the last third of the curve with the thick solid line, which causes the battery management system to wake up due to a Exceeding the limit leads and thus triggers an emergency mode of the battery management system BMS, in particular to prevent a thermal runaway of the battery system.

In the illustration in FIG. 3, the voltage U is now shown as the measured value. The situation is the following:

The starting position of the battery system is that the cell voltage U relaxes after a load. In principle, the curve can be described as the sum of e-functions, with another very small change occurring as a result of the self-discharge. If the situation is such that one of the battery cells sinks more than the remaining ones, then a critical condition may exist. However, permanent monitoring and a comparison of all battery cells is not possible, otherwise the

Battery management system BMS would have to be kept awake all the time, which is not desirable for the reasons of energy efficiency described above. So a wakeUp-capable front-end ASIC is used with the sensor arrangement accordingly. The approach to monitoring the battery system now defines a change value for the voltage U. This is denoted by ÄU _th in the representation of the figure. The following steps are then carried out: 1. Determining the lowest cell voltage for each installed sensor arrangement (ASIC).

2. Setting a voltage limit Uiim,x=U-ÄUth for the next wake-up of the battery management system BMS (wakeUp).

3. Determining the voltage gradient at time t _x .

4. Calculation of the period of time Ät _wake u _, x until the next time the battery management system BMS is woken up from the gradient and the change value ÄUth. The next regular wake-up of the battery management system BMS is then used to track the voltage limit Uiim,x.

5. The BMS battery management system is then switched back to sleep mode.

If the battery management system BMS is now woken up after the specified wake-up time Δt _wake u has elapsed, then steps 1 to 5 described are repeated accordingly. If necessary, further diagnoses and calculations are possible.

The voltage threshold value λUth is a design parameter to be specified, which can be chosen to be constant for all areas, for example. However, it is advantageously selected differently for the individual areas depending on the shape of the idling characteristic curve. Then, in areas with a flat characteristic curve, for example a state of charge of approximately 50%, a lower value would be selected, for example 5 mV, compared to higher or lower states of charge of, for example, less than 20% or more than 90%. As already mentioned above, it is used to calculate the corresponding wake-up time Δt _tW akeu using the determined voltage gradient.

The diagram in FIG. 3 shows that, for example at the end of the wake-up time Ä _tW akeu _{i ,} the distance U _re m of the actual voltage from the set voltage threshold ÄU _th is sufficiently small so that if the voltage deviates, the wake-up would take place relatively quickly . If the voltage does not deviate unexpectedly, as is the case here at points L, t ₂ and h, then, as described above, the wake-up takes place at the respective point in time at which the gradient at point L, t ₂ exceeds the corresponding voltage limit Uiim .i , m, 2 reached. During the course of the third wake-up period At _wa keu 3, the voltage drops more than expected, as shown by the solid line. The battery management system BMS is therefore woken up by a limit violation at time tn, when the actual value reaches the voltage limit Uiim,3, in this case 3.6 V. Compared to entering a standardized voltage limit of typically 2 V, the battery management system BMS wakes up much earlier, which provides the time required to be able to react in an emergency mode. Such an emergency mode is described in more detail after FIGS. 3 to 5 with its possibilities.

The illustration in FIG. 4 now shows a comparable scenario taking into account the cell temperature T, in this case when the battery system is cooling down. The starting point is that a change in temperature occurs, for example, due to a change in the ambient temperature. The cells of the battery system therefore cool down. In principle, heating or cooling of the battery system could also be a cause. It should be noted that the rate of temperature change is severely limited due to the relatively large thermal mass of the battery system.

The approach to monitoring the battery system based on the temperature T when the battery system cools down is now again based on the definition of a change value in the temperature AT _th at which a regular wake-up should take place. The procedure then includes the following steps:

1. Determination of the highest cell temperature per sensor arrangement or ASIC.

2. Setting a temperature limit T| _im ,x=T+ÄTth,max as the temperature limit for waking up in the event of a limit violation. Despite the cooling, the temperature limit Tii _m,x is higher than the current value, since the temperature T of the individual cells increases accordingly in the event of a thermal runaway.

3. Determining the temperature gradient at time t _x .

4. Calculation of the period of time until the next wake-up Ät _wa keu x based on the temperature gradient and the defined change value ÄT _t h. This regular waking up of the battery management system BMS then serves to adjust the temperature limit T _|im,x accordingly.

5. The BMS battery management system is switched back to sleep mode. The temperature value T _am b in the representation of the diagram in FIG. 4 symbolizes the ambient temperature.

The value ÄT _th.max serves to intercept a certain amount of noise in the temperature signal. In principle and theoretically, the limit could be set to the current temperature T, since the temperature T falls as time t progresses during cooling. However, if there is noise in the measured value of the temperature T, an unwanted emergency alarm is triggered very quickly, so that this offset value ÄT _{th.max leads} to a better result here. If, which is also conceivable, the value were (additionally) calculated with the defined change value ÄT _t h, cooling down with a correspondingly small temperature change could result in an extraordinarily long sleep duration, which means that waking up due to a limit violation only occurs at a very large temperature difference, which could already be critical due to the small rate of temperature change.

In the event of a wake-up after the wake-up period Δt _W akeu x has elapsed, the steps 1 to 5 mentioned are then repeated accordingly and further diagnoses and calculations are also possible here.

In the region of the third wake-up time Δt _W akeu _P 3 , temperature T breaks out again at time tn, analogously to the explanation in FIG. As soon as this temperature T reaches the corresponding temperature limit T _|im, 3 _, ie at time tn, a wake-up is initiated by a limit violation and the emergency mode of the battery management system BMS is thus initiated.

In the illustration in FIG. 5, another example is described using the temperature T as the variable to be monitored. The prerequisites here are essentially the same as in the illustration in Figure 4, except that the battery system has to be warmed up, for example due to a changing ambient temperature or a preheating of a very cold battery system, so that its performance is full from the start or at least in one is available to a sufficient extent for the planned route. The approach here is again the following. First, a change value is defined for which a regular wake-up should take place. This is in turn denoted by ÄT _th . The procedure is then largely analogous to the procedure described above, so that the following steps are carried out:

2. Setting a temperature limit Tii _m ,x=T+ ÄTth +ÄT _t h,max as a temperature limit for waking up in the event of a limit violation. Despite the cooling, the temperature limit is T| _im ,x higher than the current value, since the temperature T of the individual cells increases accordingly in the event of a thermal runaway.

3. Determining the temperature gradient at time t _x .

4. Calculation of the period of time until the next wake-up Ät _W akeu _P x based on the temperature gradient and the defined change value ÄT _t h. This regular waking up of the battery management system BMS then serves to adjust the temperature limit T _|im,x accordingly.

5. The BMS battery management system is switched back to sleep mode.

When setting the temperature limit Tii _m,x , since the direction of the temperature change when the battery system is warming up is in principle the same as in the event of overheating of one of the individual cells in accordance with the current temperature T, both the specified change value ÄT _t h and, as described above, the offset value ÄT _th.max must be added.

The rest behaves analogously. Here, too, a limit violation occurs at time tn, which wakes up the battery management system BMS and initiates an emergency mode.

For the case shown in the diagrams shown in Figures 3 to 5, in which the wake-up of the battery management system BMS has taken place through limit monitoring by a measured value of a sensor arrangement or its ASIC, an emergency must typically be assumed, i.e. for example an imminent one thermal runaway of the battery system. So if the battery management system BMS is woken up by a wake-up signal triggered by a limit being exceeded by one of the sensor systems or their ASIC, then the battery management system BMS will collect additional values, for example the insulation resistance of the battery system or individual modules as well as an impedance measurement of individual cells or individual modules of the battery system. In contrast to pure monitoring via the sensor arrangement and the ASIC, this requires active measurement mimicry, which is only available when the battery management system BMS is awake. However, it is then possible for the battery management system BMS to verify a critical state of the battery, if this is the case, on the basis of the detected variables, and to issue a warning and initiate countermeasures. If the battery is initially not in a critical state, the battery management system BMS can still be kept awake in emergency mode, for example for a fixed, predetermined period of time and can carry out further active diagnostics during this time, such as the measurements of the insulation resistance or impedance already mentioned. This ensures that a possible emergency, such as an imminent thermal runaway of the battery system, is overlooked.

If everything remains quiet for the elapse of this predetermined period of time, ie no measurements are taken which indicate a problem in the battery system, then the battery management system BMS can switch back to the idle mode after this period of time has elapsed. Further monitoring then takes place again via the sensor arrangements and their ASICs.

In addition, relative and derived quantities can be calculated from the cell voltages and temperatures recorded by the sensor arrays, which can also be used to verify a potentially critical state. This can be, for example, variances in cell voltages and temperatures, gradients in cell voltages and temperatures, and a comparison of the gradients and variances in temperatures and voltages. All of this means that, due to the closely tracked limits, the battery management system BMS is woken up extremely quickly in the event of an imminent emergency, so that there is enough time to take countermeasures such as activating or deactivating parts of the battery system, starting the cooling system, switching on or switching off consumers or the like to counteract a thermal runaway of the battery.

Claims

Mercedes-Benz Group AG patent claims

1. Method for monitoring a battery system with at least one battery module having a plurality of individual battery cells, with a battery management system (BMS) and at least one sensor arrangement (ASIC), wherein the battery management system (BMS) is not active in a sleep mode of the battery system, the sensor arrangement (ASIC) monitors at least one measured value in sleep mode and compares it with default values, in the event that the measured value deviates from the default values, the sleep mode ends and the battery management system (BMS) is activated by a wake-up signal, characterized in that the wake-up signal is additionally alternatively triggered by the expiration of a specified wake-up time (Ät _wak eu ) is triggered, whereby if the wake-up signal is triggered via the expiry of the specified wake-up time (Ät _wak eu ), a new wake-up time (Ät _wak eu ) and new default values ( _m , T _üm ) are determined and specified before the Battery management system (BMS) again in the Ru he mode changes, and in the event that the wake-up signal is triggered via a deviation of the at least one measured value from the at least one default value (Um, Tü _m ), an emergency mode is started.

2. The method according to claim 1, characterized in that in the determination of the new wake-up time (Ät _wake u) and/or the new default values (Uiim, T iim) gradients of the time course of the at least one measured value and/or a specified change (ÄT _th , ÄUt _h ) of the at least one measured value.

3. The method as claimed in claim 1 or 2, characterized in that a new wake-up time (Ät _wak eu ) and new default values (Uiim, Tn _m ) are individually determined and specified for each sensor arrangement (ASIC).

4. The method according to claim 1, 2 or 3, characterized in that to determine the new wake-up time (Ät _wake u ) and the new default values (Uiim,

T u _m ) a permitted change (ÄUt _h , ÄT _th ) of the measured value is determined on the basis of the current values and the highest and/or lowest value recorded for one of the individual battery cells by the sensor arrangement (ASIC) is determined, from which a measured value limit (Un _m , Tn _m ) is calculated for the next wake-up, and for which purpose a gradient of the measured value is calculated at the current point in time, from which a new wake-up time (ÄT _wake u ) is calculated.

5. The method as claimed in claim 4, characterized in that further diagnoses and/or calculations are carried out to validate and/or check the values determined for the new wake-up time (Ät _wak eu ) and/or the new default values (Un _m , Tn _m ). .

6. The method as claimed in one of claims 1 to 5, characterized in that the measured values monitored include at least the temperature (T) and/or the voltage (U) of the battery system or of individual battery cells.

7. The method as claimed in one of claims 1 to 6, characterized in that the measured value limit (Ui _im , Ti _im ) is calculated by adding or subtracting the defined change (ÄUth, ÄT _t h) to the current measured value.

8. The method as claimed in one of claims 1 to 7, characterized in that the sum or difference is offset against an offset value (ÄTth.tmax).

9. The method according to any one of claims 1 to 8, characterized in that the emergency mode includes further measurements and / or diagnostics by the battery management system (BMS), wherein in the event of a confirmed emergency, countermeasures are initiated and a warning is issued, and in the event of a no confirmed emergency, the battery management system (BMS) is kept awake and the at least one measured value is monitored with narrow predetermined threshold values.

10. The method as claimed in claim 9, characterized in that the emergency mode is terminated after a specified period of time has elapsed in which the threshold values have not been violated.