Method for controlling a charging current limiting value for a battery management system, battery management system

Abstract

A method for controlling a charging current limiting value for a battery management system. In one example, the method includes determining, for a measured temperature and a prescribed state of charge, reference currents for various time intervals; calculating a corresponding reference time constant for each reference current by using a model for the calculation of a mean value of a charging current based on a continuous current; constituting a diagram for the relationship between the reference time constant and the reference current; determining a predictive time constant by the comparison of a measured value of a charging current with the reference currents; calculating a predictive limiting mean value of the charging current; and calculating a first predictive limiting value ipredS for a short predictive time tpredS, a second predictive limiting value ipredL for a long predictive time tpredL, and a third predictive limiting value ipredP for a continuous predictive time.

Description

BACKGROUND OF THE INVENTION

The invention relates to a method for controlling a charging current limiting value for a battery management system.

The invention further relates to a battery management system, which is designed to execute the method according to the invention.

The invention also relates to a battery having one or more battery cells, which is designed to execute the method according to the invention and/or which comprises the battery management system according to the invention.

The invention further relates to a vehicle, which is designed to execute the method according to the invention, and/or which comprises the battery management system according to the invention and/or which comprises the battery according to the invention.

It is apparent that, in future, electrically-propelled motor vehicles will be employed to an increasing extent. In electrically-propelled motor vehicles of this type, such as e.g. hybrid vehicles and electric vehicles, rechargeable batteries are employed, predominantly for the supply of electrical energy to electric drive machines.

As an electrochemical system, a battery cell can only take up a specific amount of charge during the charging process. Under specific conditions, undesirable secondary reactions may occur, such as lithium plating, for example. During the charging process, lithium is introduced into the anode. If, however, the charging current is too high, lithium metal is deposited on the anode surface instead of being intercalated. Conditions which might lead to lithium plating are, for example, a high charging current, a long charging time without a break, a low temperature and a high state of charge SOC. Lithium plating can lead to a high loss of capacity, a significant increase in resistance and a large reduction in the performance of the battery cell. Ultimately this effect has a direct influence upon the available power, and can shorten the service life of the battery cell.

SUMMARY OF THE INVENTION

The object of the present invention is the calculation of a maximum permissible current in the charging direction without causing lithium plating.

A method is proposed for controlling a charging current limiting value for a battery management system. The battery management system is designed for the monitoring and control of a battery. The battery can comprise one or more series- or parallel-connected battery cells. The battery cell is configured in the form of a lithium-ion cell.

According to the invention, firstly, for a measured temperature T_sens and a prescribed state of charge SOC, reference currents i_ref are determined for various time intervals t_ref. The reference current i_ref represents the maximum permissible charging current at this measured temperature T_sens and this prescribed state of charge SOC for the corresponding time intervals t_ref. For example, for a measured temperature T_sens of 25° C. and a prescribed state of charge SOC, reference currents i_ref2s, i_ref10s, and i_ref30s are determined for the corresponding time intervals t_ref of 2 s, 10 s and 30 s. If, for example, the temperatures defined in the cell data sheet are from 20° C. and 30° C., these reference currents can be interpolated, if this is permitted by the cell data sheet.

Thereafter, for each reference current i_ref, a corresponding reference time constant τ_ref is calculated by the application of a model for the calculation of a mean value i_avrg of a charging current i_req, with reference to a continuous current i_cont, which corresponds to the minimum current permissible without lithium plating. The main function of the model is the authorization of a specific current for the time interval considered, by reference to the cell data sheet. This model thus permits the modelling of the dynamic behavior of current limiting values. For example, if it proceeds from the cell data sheet that a current of 150 A is only permitted to last for 2 s, this current must then be permitted for 2 s or less. To this end, the reference time constant τ_ref is adjusted such that the limiting value for current occurs at 2 s or earlier. For example, for the respective reference currents i_ref2s, i_ref10s and i_ref30s, a corresponding reference time constant τ_ref2s, τ_ref10s and τ_ref30s is calculated.

With reference to the calculated reference time constants τ_ref and the reference currents i_ref determined, a diagram is then constituted for the relationship between the reference time constant τ_ref and the reference current i_ref for each specified temperature T and each specified state of charge SOC.

Thereafter, a predictive time constant τ_pred is determined by the comparison of a measured value i_sens of a charging current i_req with the reference currents i_ref. If the measured value i_sens of the charging current i_req is equal to a reference current i_ref the predictive time constant τ_pred is equal to the reference time constant τ_ref which corresponds to this reference current i_ref. Otherwise, the predictive time constant τ_pred is determined by interpolation.

A predictive limiting mean value i_pred of the charging current i_req is then calculated on the basis of the continuous current i_cont, a predictive time t_pred and the predictive time constant τ_pred. The predictive time t_pred corresponds to those time periods during which a constant current can be employed, with no limitation executed by the battery management system. The predictive time can be customer-specific.

The reference current i_ref and the duration t_ref of the reference current i_ref are cell-specific and temperature-dependant as well as state-of-charge dependent. These variables can be sourced directly from the cell data sheet, or obtained by measurements.

The model for the calculation of the mean value i_avrg of a charging current i_req is preferably configured in the form of a PT1-element, also described as a PT1-filter. A PT1-element is understood as an LTI (linear time-invariant) transmission element in control technology, which assumes a proportional transmission behavior with a time delay of the first order. A customary example is a low-pass of the 1^st order, which can be achieved, for example, by means of a RC-element. Advantageously, a PT1 element is simple to implement.

The method according to the invention is explained by reference to the model, which is preferably configured in the form of a PT1-element. The mean value i_avrg of a charging current i_req is calculated as follows by a first equation:

$\begin{array}{cc}{i}_{avrg}\left(t\right)=\left(1-e^{-\frac{t}{\tau }}\right)·i_{sens}\left(t\right)+i_{avrg0}·e^{-\frac{t}{\tau }}& \left(1\right)\end{array}$

Where i_avrg(t) is the present mean value of the charging current i_req, and is time-dependent, i_avrg0 is the initial mean value of the charging current i_req, t is the time, t is the time constant of the PT1-element and i_sens(t) is the measured value of the charging current i_req.

The mean value i_avrg(t) of the charging current i_req is then compared as follows by means of a second equation, with a continuous current i_cont, which is temperature-dependent as well as state-of-charge dependent:i_avrg(t)≤i_cont(T,SOC)  (2)

This continuous current i_cont corresponds to the minimum current permissible without lithium plating, and can be sourced from the cell data sheet. The continuous current i_cont is relevant to the limitation of the charging current i_req, and ensures that no lithium plating of the battery cell occurs.

Provided that i_avrg(t) is equal to i_cont, a predictive limiting mean value i_pred, which corresponds to the measured value i_sens of the charging current i_req in the first equation, can be calculated as follows by means of a third equation:

$\begin{array}{cc}{i}_{\mathrm{pred}}\left(t\right)=\frac{i_{cont}\left(T,\mathrm{SOC}\right)-i_{avrg0}·e^{-\frac{t}{\tau }}}{1-e^{-\frac{t}{\tau }}}& \left(3\right)\end{array}$

Where T is the temperature and SOC is the state of charge of the battery cell.

For a specific temperature T and a specific predictive time interval t_pred, a new third equation can be derived as follows:

$\begin{array}{cc}{i}_{\mathrm{pred}}\left(t_{\mathrm{pred}}\right)=\frac{i_{\mathrm{cont}}-i_{\mathrm{avrg}0}·e^{-\frac{t_{\mathrm{pred}}}{\tau }}}{1-e^{-\frac{t_{\mathrm{pred}}}{\tau }}}& \left(3‘\right)\end{array}$

In this case, a constant k is introduced. This constant k is defined as follows:

$k=e^{-\frac{t_{\mathrm{pred}}}{\tau }}$

A simplified new third equation is thus derived from the new third equation as follows:

$\begin{array}{cc}{i}_{\mathrm{pred}}\left(t_{\mathrm{pred}}\right)=\frac{i_{\mathrm{cont}}-i_{\mathrm{avrg}0}·k}{1-k}& \left(3‘‘\right)\end{array}$

The initial mean value i_avrg0 can be obtained by means of the reference current i_ref and the time t_ref of the reference current i_ref, in the form of a step response, which is described as follows in a fourth equation:

$\begin{array}{cc}{i}_{avrg0}\left(t_{\mathrm{ref}}\right)=\left(1-e^{-\frac{t_{ref}}{\tau }}\right)·i_{\mathrm{ref}}& \left(4\right)\end{array}$

With reference to the simplified new third equation (3″) and the fourth equation, the predictive limiting mean value i_pred for a specific predictive time interval t_pred is calculated as follows by means of a fifth equation:

$\begin{array}{cc}{i}_{\mathrm{pred}}\left(t_{\mathrm{pred}}\right)=\frac{i_{cont}-\left(1-e^{-\frac{t_{ref}}{\tau }}\right)·i_{\mathrm{ref}}·k}{1-k}& \left(5\right)\end{array}$

Provided that i_pred (t_pred) is equal to i_ref, the time constant t is derived as follows by means of a sixth equation:

$\begin{array}{cc}\tau =\frac{-t_{\mathrm{ref}}}{\ln \left(\frac{i_{\mathrm{ref}}-i_{cont}}{k·i_{ref}}\right)}& \left(6\right)\end{array}$

A time constant τ is thus calculated for each specified reference current i_ref and the specified time t_ref. By means of the sixth equation, a new time constant t can be calculated for each specified temperature T and each specified state of charge SOC, which is adjusted according to the measured value i_sens of the charging current i_req.

The measured value i_sens of the charging current i_req is compared with the reference current i_ref for a specific temperature T, a specific state of charge SOC and a specific time t_ref, in order to derive an appropriate predictive time constant τ_pred.

If, for example, the measured value i_sens of the charging current i_req is equal to the reference current i_ref2s for a time t_ref of 2 s, a predictive time constant τ_pred can be calculated by means of the sixth equation. The predictive time constant τ_pred is equal to the reference time constant τ_ref2s for the reference current i_ref2s for a time t_ref of 2 s. The reduction of the charging current i_req then commences after 2 s.

Determination of the predictive time constant τ_pred is preferably executed by linear interpolation. The predictive time constant τ_pred is thus dynamic, and is adjusted according to the measured value i_sens of the charging current i_req.

If, for example, the measured value i_sens of the charging current i_req is greater than the reference current i_ref10s for a time t_ref of 10 s, but is smaller than the reference current i_ref2s for a time t_ref of 2 s, a predictive time constant τ_pred can be determined by linear interpolation between the reference time constant τ_ref2s for the reference current i_ref2s for a time t_ref of 2 s and the reference time constant τ_ref10s for the reference current i_ref10s for a time t_ref of 10 s.

By means of the predictive time constant τ_pred thus determined and the third equation, a predictive limiting mean value i_pred can be calculated.

Preferably, an additional point [i_min; τ_relax] is inserted in the diagram for the relationship between the reference time constant τ_ref and the reference current i_ref. This point is inserted in order to define a relaxation time constant τ_relax for the battery cell in a relaxed or quasi-relaxed state. Thus, i_min represents a small current, at which no loading of the cell occurs. This small current i_min can be a small charging current or discharge current. By means of this definition, a small relaxation time constant τ_relax can be selected, for example, in order to permit a high recuperation current. This new point can be dependent upon the temperature T and the state of charge SOC.

Preferably, on the basis of the limiting mean value i_pred, a first predictive limiting value i_predS for a short predictive time t_predS, a second predictive limiting value i_predL for a long predictive time t_predL and a third predictive limiting value i_predP for a continuous predictive time t_predP are calculated. For example, a time of less than 2 s can be defined as a short predictive time t_predS. For example, a long predictive time t_predL can be equal to 2 s, whereas a continuous predictive time t_predP can be equal to 10 s. These predictive times can be customer-specific. In general, these predictive times are fixed values. The predictive time t_pred corresponds to those time periods during which a constant current can be employed, with no limitation executed by the battery management system. This constant current can be employed, for example, for the calculation of the third predictive limiting value i_predP.

Preferably, the constant

$k=e^{-\frac{t_{pred}}{\tau }}$ employed in the calculation of the first predictive limiting value i_predS is established. By the establishment of the constant k and the variation of the time constant τ, which is also described as a short predictive time constant τ_predS, the first predictive limiting value i_predS is also dynamic, as it adjusts according to the present thermal loading and the loading for lithium plating of the battery cell.

For the calculation of the second predictive limiting value i_predL and a third predictive limiting value i_predP, two conditions are to be fulfilled.

Firstly:k_S≥k_L(T,SOC)≥k_P(T,SOC)

Where ks is the constant

$k=e^{-\frac{t_{pred}}{\tau }}$ for the first predictive limiting value i_predS, which is a fixed value, k_L(T, SOC) is the constant

$k=e^{-\frac{t_{pred}}{\tau }}$ for the second predictive limiting value i_predL and k_P(T, SOC) is the constant

$k=e^{-\frac{t_{pred}}{\tau }}$ for the third predictive limiting value i_predP. The constants k_L(T, SOC) and k_P(T, SOC) are temperature-dependent as well as state-of-charge dependent.

Secondly, if the PT1-element is equal to zero, it is the case that the predictive limiting value i_predL for a long predictive time t_predL should represent the reference current i_ref for this long time t_predL and for the present temperature T of the battery cell. By “PT1-element is equal to zero”, it is to be understood that the mean value I_avrg of the charging current i_req is equal to zero. The constant k_L(T, SOC) for the second predictive limiting value i_predL is defined as follows:

$k_L\left(T,\mathrm{SOC}\right)=e^{-\frac{t_{pr\mathrm{edL}}}{{{\tau }_{\mathrm{predL}\left(T,\mathrm{SOC}\right)}}^{}}}$

In consideration of these two conditions, the constant k_L and a long predictive time constant τ_predL(T, SOC) for the second predictive limiting value i_predL are calculated as follows by means of a seventh and eighth equation:

$\begin{array}{cc}{k}_L\left(T,\mathrm{SOC}\right)=1-\frac{i_{cont}\left(T,\mathrm{SOC}\right)}{i_{\mathrm{ref},\mathrm{predL}}\left(T,\mathrm{SOC}\right)}& \left(7\right)\end{array}$ $\begin{array}{cc}{\tau }_{\mathrm{predL}}\left(T,\mathrm{SOC}\right)=-\frac{t_{predL}}{\ln \left(k_L\left(T,\mathrm{SOC}\right)\right)}& \left(8\right)\end{array}$

Where i_cont(T, SOC) is the continuous current corresponding to the minimum current permissible without lithium plating, and can be sourced from the cell data sheet. i_ref,predL is the reference current i_ref for this long predictive time t_predL.

The second predictive limiting value i_predL is then calculated as follows by means of a ninth equation, which is derived from the third equation:

$\begin{array}{cc}{i}_{\mathrm{predL}}\left(t\right)=\frac{i_{cont}\left(T,\mathrm{SOC}\right)-i_{avrg0}·e^{-\frac{t_{predL}}{{\tau }_{predL}\left(T,\mathrm{SOC}\right)}}}{1-e^{-\frac{t_{predL}}{{\tau }_{predL}\left(T,\mathrm{SOC}\right)}}}& \left(9\right)\end{array}$

Calculation of the third predictive limiting value i_predP is subject to the same conditions as those applied to the calculation of the second predictive limiting value i_predL. Thus, k_L(T,SOC)≥k_P(T,SOC). Accordingly, the third predictive limiting value i_predP can be calculated as follows by means of a tenth equation:

$\begin{array}{cc}{i}_{\mathrm{predP}}\left(t\right)=\frac{i_{cont}\left(T,\mathrm{SOC}\right)-i_{avrg0}·e^{-\frac{t_{p\mathrm{redP}}}{{\tau }_{p\mathrm{redP}}\left(T,\mathrm{SOC}\right)}}}{1-e^{-\frac{t_{predL}}{{\tau }_{p\mathrm{redP}}\left(T,\mathrm{SOC}\right)}}}& \left(10\right)\end{array}$

Where τ_predP(T, SOC) is the continuous predictive time constant for the third predictive limiting value i_predP.

A battery management system is further proposed, which is designed to execute the method according to the invention.

A battery having one or more battery cells is also proposed, which is designed to execute the method according to the invention, and/or which comprises the battery management system according to the invention.

A vehicle is further proposed, which is designed to execute the method according to the invention, and/or which comprises the battery management system according to the invention and/or the battery according to the invention.

Advantages of the Invention

By means of the present invention, an algorithm can be efficiently implemented in a battery management system, which limits the current in accordance with the cell specifications, and thus ensures the full availability of capacity and the protection of battery cells against lithium plating and components.

Data from battery cells are directly employed for the calculation of a time constant. By means of the method according to the invention, the predictive limiting mean value of the charging current and the predictive limiting values of the charging current for different predictive times can be calculated. These variables are advantageous, firstly for the detection of potential premature ageing in battery cells, and secondly for the adjustment of a power strategy and a driving strategy for a vehicle.

The method according to the invention moreover permits the determination of a stress level, for the quantification of cell utilization and rapid calibration.

The method according to the invention moreover permits the calculation of the mean value i_avrg of the charging current i_req, which can be employed as a measure of battery loading. This mean value i_avrg of the charging current i_req can be recorded, as a means of visualizing the stress level of the cell.

Moreover, implementation of the method according to the invention is efficient, as only limited RAM is required for the processors of the battery management system. The method according to the invention takes account of the stress experienced by the cell, using a single integrator.

The method is also flexible in operation. The predictive time horizon or predictive time interval can thus be varied, according to the requirements of a customer.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are described in greater detail with reference to the drawings and the following description.

In the drawings:

FIG. 1 shows a schematic representation of the anticipated behavior of a charging current limiting value,

FIG. 2 shows a schematic representation of a diagram for determining a predictive time constant τ_pred,

FIG. 3A shows a schematic representation of a temporal characteristic of a limiting mean value i_pred,

FIG. 3B shows a schematic representation of a temporal characteristic of a mean value i_avrg of the charging current i_req according to FIG. 3A,

FIG. 3C shows a schematic representation of a temporal characteristic of the predictive time constant τ_pred according to FIG. 3A,

FIG. 4A shows a schematic representation of a temporal characteristic of a charging current i_req,

FIG. 4B shows a schematic representation of a temporal characteristic of a measured voltage u_sens of the battery cell according to FIG. 4A,

FIG. 4C shows a schematic representation of a temperature characteristic of a charging current i_req with limitation of the mean value i_avrg thereof according to FIG. 4A,

FIG. 5A shows a schematic representation of a temporal characteristic of a charging current i_req, with limitation of the mean value i_avrg thereof according to a first example,

FIG. 5B shows a schematic representation of a temporal characteristic of a predictive time constant τ_pred according to FIG. 5A,

FIG. 5C shows a schematic representation of a temporal characteristic of a state of charge SOC and a temporal characteristic of a measured temperature T_sens according to FIG. 5A, FIG. 6A shows a schematic representation of a temporal characteristic of a charging current i_req, with limitation of the mean value i_avrg thereof, according to a second example,

FIG. 6B shows a schematic representation of a temporal characteristic of a predictive time constant τ_pred according to FIG. 6A,

FIG. 6C shows a schematic representation of a temporal characteristic of a state of charge SOC and a temporal characteristic of a measured temperature T_sens according to FIG. 6A, and

FIG. 7 shows a sequence for the method according to the invention.

In the following description of embodiments of the invention, identical or similar elements are identified by the same reference symbols, wherein any repeated description of these elements in individual cases is omitted. The figures represent the subject matter of the invention in a schematic manner only.

DETAILED DESCRIPTION

FIG. 1 shows a schematic representation of the anticipated behavior of a charging current limiting value of a battery cell. It is anticipated that, by the employment of dynamic limiting values in in a battery management system for the monitoring and control of the battery cell, the initial value of a charging current i_req, in a first phase 12 of duration, for example, 30s, is not reduced, and these dynamic limiting values i_D are converged in a second phase 14 thereafter to constitute continuous limiting values i_C. In a third phase 16, the cell current i_req is then limited by the continuous limiting values i_C.

FIG. 2 shows a schematic representation of a diagram for determining a predictive time constant τ_pred. This diagram is clarified hereinafter with reference to a model for the calculation of a mean value i_avrg of a charging current i_req using a PT1-element. As described above, a time constant τ is calculated for each specific reference current i_ref, a specific time interval t_ref and a specific temperature T and a specific state of charge SOC. In the present case, in FIG. 2, for a specific temperature T, a reference time constant τ_ref30s for a reference current i_ref30s of duration 30s, a reference time constant τ_ref10s for a reference current i_ref10s of duration 10 s and a reference time constant τ_ref2s for a reference current i_ref2s of duration 2 s are calculated.

By means of these data, a diagram is plotted for the relationship between the time constant t and the reference current i_ref in FIG. 2.

The measured value i_sens of the charging current i_req is compared with the reference current i_ref for a specific time interval t_ref, in order to derive an appropriate predictive time constant τ_pred.

If, for example, the measured value i_sens of the charging current i_req is equal to the reference current i_ref2s, a predictive time constant τ_pred is calculated which is equal to the reference time constant τ_ref2s which has been calculated for the reference current i_ref2s. The reduction of the charging current i_req then commences after 2 s.

If, for example, the measured value i_sens of the charging current i_req is greater than the reference current i_ref10s, but is smaller than the reference current i_ref2s, a predictive time constant τ_pred is determined by linear interpolation between the reference time constant τ_ref10s and the reference time constant τ_ref2s.

Moreover, in the diagram according to FIG. 2, an additional point [i_min; τ_relax] is inserted. This point is inserted, in order to define a relaxation time constant τ_relax for the battery cell in a relaxed or quasi-relaxed state. Thus, i_min represents a small current. By means of this definition, a small relaxation time constant τ_relax can be selected in order to permit, for example, a high recuperation current. This new point can thus be dependent upon the temperature T and the state of charge SOC.

FIG. 3A shows a schematic representation of a temporal characteristic of a limiting mean value i_pred. A measured value i_sens of the charging current i_req of 60 A is detected. A charging current i_req of 60 A is only permissible for a time of 2 s, without causing lithium plating. A limiting mean value i_pred is thus calculated by the method proposed according to the invention. Reduction of the charging current i_req then commences after 2 s. The limiting mean value i_pred ultimately converges to a continuous current i_cont, which corresponds to the maximum permissible continuous charging current i_req.

FIG. 3B shows a schematic representation of a temporal characteristic of a mean value i_avg of the charging current i_req according to FIG. 3A, whereas FIG. 3C shows a schematic representation of a temporal characteristic of a predictive time constant τ_pred according to FIG. 3A. From FIG. 3C, it can be seen that the predictive time constant τ_pred is adjusted according to the measured value i_sens of the charging current i_req.

FIG. 4A shows a schematic representation of a temporal characteristic of a charging current i_req. The charging current i_req is pulse-shaped, and comprises two current pulses with equal measured values i_sens of 60 A. The duration of the respective current pulses is 2 s. At time t₁, a first current pulse is transmitted, and the first current pulse ends at time t₂. At time t₃, a second current pulse is transmitted, and the second current pulse ends at time t₄.

FIG. 4B shows a schematic representation of a temporal characteristic of a measured voltage u_sens of the battery cell according to FIG. 4A. The measured voltage u_sens of the battery cell has a no-load voltage u_OCV at first. Charging with the charging current i_req increases the measured voltage u_sens of the battery cell. The measured voltage u_sens of the battery cell decreases only from the time t₂. In an intermediate time period t_relax between the first and the second current pulse, specifically between the time t₂ and the time t₃, the measured voltage u_sens decreases at a rate of, for example, 1 mV/min to the no-load voltage u_OCV. This intermediate time period t_relax is also described as the relaxation time.

FIG. 4C shows a schematic representation of a temporal characteristic of a charging current i_req, with the limitation of the mean value i_avrg thereof. The calculated limiting mean value i_pred also rises during the intermediate time period t_relax or the relaxation time, in order to allow the second current pulse. A cell must be stress-relieved or relaxed, before a further current pulse can be delivered at the maximum permissible capacity. In a resting cell, the measured voltage corresponds to the no-load voltage u_OCV. For this reason, it is important that a sufficiently long relaxation time should be incorporated, in order to permit the second current pulse. This relaxation time corresponds to the time required for the measured voltage to achieve the no-load voltage of the cell. It will then be possible to set the maximum power, with no risk of lithium plating. This parameter can vary, according to the temperature T, the state of charge SOC and the current strength of the previously employed pulse.

FIG. 5A shows a schematic representation of a temporal characteristic of a charging current i_req, with limitation of the mean value i_avrg thereof, according to a first example, whereas FIG. 5B shows a schematic representation of a temporal characteristic of a predictive time constant τ_pred according to FIG. 5A, and FIG. 5C shows a schematic representation of a temporal characteristic of a state of charge SOC and a temporal characteristic of a measured temperature T_sens according to FIG. 5A.

Temporal characteristics of a relaxed battery cell are represented having an initial state of charge SOC of 85%. An initial temperature T of the battery cell is −10° C. The battery cell is thus charged with a charging current i_req of 175 A for a time of 30s. The state of charge SOC and the measured temperature T_sens remain unchanged.

From FIG. 5A, it can be seen that, at time point t=10 s, a first current pulse, which represents the charging current i_req, having a measured value i_sens of 175 A, is transmitted to the battery cell. The duration of the first current pulse is 30s. From the data sheet for the battery cell 34, it can be determined that a current pulse of 175 A at a temperature T of −10° C. and a state of charge SOC of 85% is only permissible for 10 s. A predictive time constant τ_pred and a limiting mean value i_pred, which converge to a continuous current i_cont, are calculated. From FIG. 5A, it can further be seen that, at time point t=20 s, i.e. after 10 s following the transmission of the first current pulse, the reduction of the first current pulse commences. The first current pulse is reduced to the continuous current i_cont. Only at the end of the first current pulse does the limiting mean value i_pred begin to rise again, in order to permit a further current pulse. At time point t=100s, a second current pulse, which is equal to the first current pulse, is transmitted to the battery cell. Given the loaded state of the battery cell, reduction of the second current pulse commences earlier.

FIG. 6A shows a schematic representation of a temporal characteristic of a charging current i_req, with limitation of the mean value i_avrg thereof, according to a second example, whereas FIG. 6B shows a schematic representation of a temporal characteristic of a predictive time constant τ_pred according to FIG. 6A, and FIG. 6C shows a schematic representation of a temporal characteristic of a state of charge SOC and a temporal characteristic of a measured temperature T_sens according to FIG. 6A.

The temporal characteristics are represented for a relaxed battery cell having an initial state of charge SOC of 85%. An initial temperature T of the battery cell is −10° C. The battery cell is charged with a charging current i_req of 175 A for a time of 30s. The state of charge SOC remains unchanged, whereas the measured temperature T_sens rises during the duration of the current pulse.

From FIG. 6A, it can be seen that, at time point t=10 s, a current pulse, which represents the charging current i_req, having a measured value i_sens of 175 A, is transmitted to the battery cell. The duration of the current pulse is 30s. From the data sheet for the battery cell, it can be determined that a current pulse of 175 A at a temperature T of −10° C. and a state of charge SOC of 85% is only permissible for 10 s. A predictive time constant τ_pred and a limiting mean value i_pred, which converges to a continuous current i_cont, are calculated. As the measured temperature T_sens of the battery cell varies over the duration of the current pulse, the predictive time constant τ_pred is calculated dynamically. From FIG. 6A, it can further be seen that the reduction of the current pulse commences somewhat later. The current pulse reduces to the continuous current i_cont. The continuous current i_cont also adjusts to the temperature T.

FIG. 7 shows a sequence for the method according to the invention. In a step S1, for a measured temperature T_sens and a prescribed state of charge SOC, reference currents i_ref are determined for various time intervals t_ref. For example, for a measured temperature T_sens of 25° C. and a prescribed state of charge SOC, reference currents i_ref2s, i_ref10s, i_ref30s are determined for the corresponding time intervals t_ref of 2 s, 10 s and 30s. If, for example, the temperatures T defined in the cell data sheet are from 20° C. and 30° C., these reference currents i_ref can be interpolated, if this is permitted by the cell data sheet.

In a step S2, for each reference current i_ref, a corresponding reference time constant τ_ref is calculated by the application of a model for the calculation of a mean value i_avrg of a charging current i_req with reference to a continuous current i_cont, which corresponds to the minimum current permissible without lithium plating. For example, if it proceeds from the cell data sheet that a current of 150 A is only permitted to last for 2 s, this current must then be permitted for 2 s or less. To this end, the reference time constant τ_ref is adjusted such that the limiting value for current occurs at 2 s or earlier. For example, for the respective reference currents i_ref2s, i_ref10s and i_ref30s, a corresponding reference time constant τ_ref2s, τ_ref10s and τ_ref30s is calculated. The model is preferably configured in the form of a PT1-element.

In a step S3, with reference to the calculated reference time constants τ_ref and the reference currents i_ref determined, a diagram is constituted for the relationship between the reference time constant τ_ref and the reference current i_ref for each specified temperature T and each specified state of charge SOC.

In a step S4, a predictive time constant τ_pred is determined by the comparison of a measured value i_sens of a charging current i_req with the reference currents i_ref. If the measured value i_sens of the charging current i_req is equal to a reference current i_ref, the predictive time constant τ_pred is equal to the reference time constant t_ref which corresponds to this reference current i_ref. Otherwise, the predictive time constant τ_pred is determined by interpolation.

In a step S5, a predictive limiting mean value i_pred of the charging current i_req is calculated on the basis of the continuous current i_cont, a predictive time t_pred and the predictive time constant τ_pred. The predictive time can be customer-specific.

In a step S6, on the basis of the limiting mean value i_pred, a first predictive limiting value i_predS for a short predictive time t_predS, a second predictive limiting value i_predL for a long predictive time t_predL and a third predictive limiting value i_predP for a continuous predictive time t_predP are calculated. For example, a time of less than 2 s can be defined as a short predictive time t_predS. For example, a long predictive time t_predL can be equal to 2 s, whereas a continuous predictive time t_predP can be equal to 10 s.

The invention is not limited to the exemplary embodiments described herein and the aspects thereof indicated. Instead, within the field indicated by the claims, a plurality of variations are possible, which lie within the practice of a person skilled in the art.

Claims

1. A method for controlling a charging current limiting value for a battery management system, comprising the following steps: Determination, for a measured temperature Tsens and a prescribed state of charge SOC, of reference currents iref for various time intervals tref;Calculation of a corresponding reference time constant τref for each reference current iref by the application of a model for the calculation of a mean value iavrg of a charging current ireq with reference to a continuous current icont, which corresponds to the minimum current permissible without lithium plating;Constitution of a diagram for the relationship between the reference time constant τref and the reference current iref, with reference to the calculated reference time constants τref and the reference currents iref determined for each specific temperature T and each specific state of charge SOC;Determination of a predictive time constant τpred by the comparison of a measured value isens of a charging current ireq with the reference currents iref; andCalculation of a predictive limiting mean value ipred of the charging current ireq, on the basis of the continuous current icont, a predictive time tpred and the predictive time constant τpred.

2. The method according to claim 1, wherein the model for the calculation of a mean value iavrg of the charging current ireq is configured in the form of a PT1-element.

3. The method according to claim 2, wherein an additional point [imin; τrelax] is inserted in the diagram for the relationship between the reference time constant τref and the reference current iref.

4. The method according to claim 1, wherein the calculation of the predictive time constant τpred corresponding to the measured value isens of the charging current ireq is executed by linear interpolation.

5. The method according to claim 1, wherein, a first predictive limiting value ipredS for a short predictive time tpredS, a second predictive limiting value ipredL for a long predictive time tpredL, and a third predictive limiting value ipredP for a continuous predictive time tpredP are calculated.

6. The method according to claim 5, wherein a constant k=e tpred/τ is defined in the calculation of the first predictive limiting value ipredS.

7. A battery management system configured to control a charging current limiting value for a battery management system, by: determining, for a measured temperature Tsens and a prescribed state of charge SOC, reference currents iref for various time intervals tref;calculating a corresponding reference time constant τref for each reference current iref by the application of a model for the calculation of a mean value iavrg of a charging current ireq with reference to a continuous current icont, which corresponds to the minimum current permissible without lithium plating;constituting a diagram for the relationship between the reference time constant τref and the reference current iref, with reference to the calculated reference time constants τref and the reference currents iref determined for each specific temperature T and each specific state of charge SOC;determining a predictive time constant τpred by the comparison of a measured value isens of a charging current ireq with the reference currents iref; andcalculating a predictive limiting mean value ipred of the charging current ireq, on the basis of the continuous current icont, a predictive time tpred and the predictive time constant τpred.

8. A battery having one or more battery cells, which is configured to control a charging current limiting value for a battery management system, by: determining, for a measured temperature Tsens and a prescribed state of charge SOC, reference currents iref for various time intervals tref;calculating a corresponding reference time constant τref for each reference current iref by the application of a model for the calculation of a mean value iavrg of a charging current ireq with reference to a continuous current icont, which corresponds to the minimum current permissible without lithium plating;constituting a diagram for the relationship between the reference time constant τref and the reference current iref, with reference to the calculated reference time constants τref and the reference currents iref determined for each specific temperature T and each specific state of charge SOC;determining a predictive time constant τpred by the comparison of a measured value isens of a charging current ireq with the reference currents iref; andcalculating a predictive limiting mean value ipred of the charging current ireq, on the basis of the continuous current icont, a predictive time tpred and the predictive time constant τpred.

9. A vehicle, which comprises a battery management system configured to control a charging current limiting value for a battery management system, by: determining, for a measured temperature Tsens and a prescribed state of charge SOC, reference currents iref for various time intervals tref;calculating a corresponding reference time constant τref for each reference current iref by the application of a model for the calculation of a mean value iavrg of a charging current ireq with reference to a continuous current icont, which corresponds to the minimum current permissible without lithium plating;constituting a diagram for the relationship between the reference time constant τref and the reference current iref, with reference to the calculated reference time constants τref and the reference currents iref determined for each specific temperature T and each specific state of charge SOC;determining a predictive time constant τpred by the comparison of a measured value isens of a charging current ireq with the reference currents iref; andcalculating a predictive limiting mean value ipred of the charging current ireq, on the basis of the continuous current icont, a predictive time tpred and the predictive time constant τpred.