DE102017218715A1 - Determination of SOC and temperature of a lithium-ion cell by means of impedance spectroscopy - Google Patents

Abstract

The invention relates to a method for determining temperature T and charge state SOC of a lithium-ion cell, the measurement of at least two impedance values A and B, the provision of reference data for A and B as a function of T and SOC, and the determination of T and SOC by Comparison of the measured impedance values A and B with the reference data includes. Another aspect of the invention relates to an apparatus for carrying out the method.

Description

Technical area

The present invention relates to a method for determining the state of charge (SOC) and the temperature of a lithium-ion cell by means of impedance spectroscopy, a device, in particular a control device for cell monitoring or battery management, which uses the method, as well as a battery system, the device includes.

Technical background

Electrochemical impedance spectroscopy (EIS) is an established method for the characterization of electrochemical systems, in particular also galvanic cells, which generally involves the measurement of the impedance, ie of the complex alternating current resistance, as a function of the frequency of the excitation signal.

It is known in the art to use impedance measurements or impedance spectroscopy to diagnose the state of lithium-ion cells.

This is for example on the US 2009/0096459 which describes a method of measuring the frequency dependence of the impedance to determine the mobility of the charges on the surface of the negative and positive electrodes of a lithium ion cell. In doing so, special methods are used to create equivalent circuit diagrams and determine the parameters for modeling cell behavior. Intended use is in particular the quality control during battery production.

DE 10 2009 038 663 describes a motor vehicle with a plurality of batteries, each separated individually or in blocks from the electrical system and can be connected to a diagnostic device for performing a model-based battery diagnostic method. The diagnostic method may also include a measurement of the impedance.

Furthermore, it is also known to determine the temperature of a lithium-ion cell by means of the impedance. Compared to a direct temperature measurement by means of a sensor on or in the cell housing, this has the advantage that the temperature in the electrochemically active region of the electrode assembly is determined directly, which may differ from the housing temperature.

DE 10 2013 103 921 relates to cell temperature measurement and degradation measurement in lithium battery systems of electrically powered vehicles by determining the cell impedance based on an AC signal given by an inverter. The method is based on the observation that the course of the application of impedance versus signal frequency is temperature-dependent.

EP 2 667 166 A2 relates to a method for determining the temperature by measuring the imaginary part of the impedance at a plurality of frequencies and determining the frequency at which the imaginary part has a zero crossing. The method is based on the observation that the frequency of the zero crossing at given charge and aging state of the cell depends essentially on the temperature.

US 2013/0264999 relates to a battery charging system that includes a temperature sensor that is alternately connected in a time division manner to the individual cells to be charged in order to measure the impedance of the cell and to determine the temperature from the phase of the impedance. The temporal temperature change rate serves as an indicator of whether the cell has been fully charged.

task

The previously known methods for determining the temperature by means of impedance measurement have the disadvantage that the frequency dependence or the phase of the impedance depends not only on the temperature but also on the state of charge (SOC) of the cell. For an accurate temperature measurement, it is therefore necessary to additionally know the state of charge, which is typically calculated by the battery control system based on the measured current and voltage data and possibly also from the (housing) temperature. Thus, the accuracy of temperature determination via the impedance in the prior art depends on how precisely the SOC can be determined.

In view of this, a method would be desirable with which the temperature can be determined directly by means of impedance measurement, without knowing the SOC. Furthermore, a method by which the SOC can be directly determined without knowing the history of the cell and the charge that has already flowed would also be desirable.

Summary of the invention

1. (a) Messung des Realteils oder des Betrages der Impedanz Z der Zelle bei einer Frequenz f1, um einen ersten Impedanzwert A zu erhalten;
2. (b) Messung mindestens eines weiteren Impedanzwerts B, ausgewählt aus:
   • - Imaginärteil bzw. Phase der Impedanz Z bei der Frequenz f1
   • - Realteil oder Betrag der Impedanz Z bei einer anderen Frequenz f2, die so gewählt ist, dass sich die T-Abhängigkeit und/oder die SOC-Abhängigkeit der Impedanz von derjenigen bei f1 unterscheidet;
   • - Imaginärteil bzw. Phase der Impedanz Z bei der Frequenz f2;
3. (c) Bereitstellung von Impedanz-Referenzdaten für die in (a) und (b) gemessenen Impedanzwerte als Funktion von T und SOC, A_ref (T, SOC) und B_ref (T, SOC), wobei die Impedanz-Referenzdaten aus vorbekannten zellspezifischen Kalibrierungsdaten abgelesen, interpoliert oder berechnet werden;
4. (d) Bestimmung von T und SOC aus den in a und b gemessenen Impedanzwerten A und B durch Vergleich mit den Referenzdaten, wobei T und SOC diejenigen sind, bei denen ΔA = A-A_ref (T, SOC) und ΔB=B-B_ref (T, SOC) gleichzeitig minimal werden.

The present invention relates to a method for determining temperature T and state of charge SOC of a lithium-ion cell, comprising:

1. (a) measuring the real part or the magnitude of the impedance Z of the cell at a frequency f1 to obtain a first impedance value A;
2. (b) measuring at least one further impedance value B selected from:
   • - Imaginärteil or phase of the impedance Z at the frequency f1
   • The real part or magnitude of the impedance Z at a different frequency f2 chosen so that the T-dependence and / or the SOC-dependence of the impedance differs from that at f1;
   • Imaginary part or phase of the impedance Z at the frequency f2;
3. (c) providing impedance reference data for the impedance values measured in (a) and (b) as a function of T and SOC, A _ref (T, SOC) and B _ref (T, SOC), the impedance reference data being from previously known cell-specific calibration data are read, interpolated or calculated;
4. (d) Determination of T and SOC from the impedance values A and B measured in a and b by comparison with the reference data, where T and SOC are those where ΔA = AA _ref (T, SOC) and ΔB = BB _ref (T , SOC) at the same time become minimal.

Another aspect of the invention relates to a battery control device which employs the method.

list of figures

• 1 shows a plot of the real part of the impedance Re (Z) (mΩ) of a lithium-ion cell against the temperature T (° C) and the state of charge SOC (%). The plot of the hatched area plot at Re (Z) = 0.79 mΩ parallel to the T-SOC plane provides an isoline of possible temperature / SOC value pairs corresponding to this Re (Z).
• 2 shows a corresponding plot of the imaginary part of the impedance Im (Z) (10 ^-4 Ω) of a lithium-ion cell against the temperature T (° C) and the state of charge SOC (%). The intersection of the plot with the hatched area at Im (Z) = 0.69 mΩ parallel to the T-SOC plane provides an isoline of possible temperature / SOC value pairs corresponding to this Im (Z).
• 3 shows the projection of the two isolines with Re (Z) = 0.79 mΩ and Im (Z) = 0.69 mΩ 1 and 2 to the T-SOC level. Re (Z) iso line covers a temperature interval ΔT of about 18 ° C in the range of 0 to 100% SOC, while the Im (Z) line covers a temperature interval ΔT of 12 ° C. As can be seen, however, there is only one point with Z = 0.79 mΩ + i * 0.69 mΩ, and this is at 25 ° C and 70% SOC.
• 4 shows schematically the course of the method according to the invention.

Detailed description

principle of operation

As described above, the impedance is generally dependent on both temperature and SOC.

Thus, for example, at constant SOC for the real part of the impedance Re (Z) different values are obtained with varying temperature. Likewise, different values are obtained for Re (Z) at constant temperature and varying SOC. For a certain, fixed Re (Z) value, there is a bevy of different temperature and SOC value pairs. In a plot of Re (Z) against temperature T and SOC, for a constant Re (Z), we find an isoline that does not run parallel to the axis of the T and SOC axes (see 1 ). Thus, T can only be determined from a measured value for Re (Z), although the SOC is known, and vice versa. Otherwise, depending on the SOC, the isoline of Re (Z) would cover a certain temperature range ΔT, which may be, for example, about 18 ° C, as in 3 shown.

The same applies to the imaginary part Im (Z) of the impedance. A plot of Im (Z) versus T and SOC also yields an isoline for each value of Im (Z), which is normally not axis-parallel to the T and SOC axes (see 2 ). Therefore, depending on the SOC, the isoline of Im (Z) also covers a certain temperature range ΔT, for example about 12 ° C, as in 3 shown.

The extent of the SOC or T-dependence of the impedance is frequency-dependent. The capacitive and diffusion-controlled contributions to impedance to which the influence of T and SOC is expected to be strongest decrease with increasing frequency, so that at high frequencies both the SOC and T dependencies decrease. However, the SOC dependence decreases significantly more with increasing frequency than the T-dependence.

One aspect of the present invention is based on the observation that dependence of the impedance on the temperature and the SOC at a given frequency F1 is typically different for the real part Re (Z) and the imaginary part Im (Z) and may also have a different course ,

While for each measured Re (Z) or Im (Z) there is a whole family of different T and SOC value pairs leading to that reading, it has been found that for a measured Re (Z) Im (Z ) Value pairs typically only have one associated T and SOC value pair. When plotting the real or imaginary part against T and SOC, the isolines for a measured value pair intersect at the point of projection into the T-SOC plane at a point corresponding to the associated T-SOC value pair (see 3 ).

This principle is, of course, independent of how the impedance is represented, and thus applies not only to the display with real and imaginary part, but also applies accordingly to the representation with magnitude and phase. It is only important that the impedance is detected as a complex variable.

The same applies analogously to the T or SOC dependence of the impedance at a different frequency f2, the course of which likewise typically differs from that at the frequency f1, the SOC dependence decreasing more rapidly with increasing frequency than the T dependence. The projections of the isolines into the T-SOC plane therefore become progressively flatter with increasing frequency, and the ΔT values covered in the range of 0 to 100% SOC decrease.

Another aspect of the present invention takes advantage of this difference. For the real part or the magnitude of the impedance at the frequency f2, it is likewise possible to determine an isoline of T-SOC value pairs which likewise typically intersects at one point with that at the frequency f1, so that in this way too a unique value can be determined for T and SOC.

Overall, therefore, the present invention is based on the measurement of at least two impedance values A and B, which are known to have a different T and SOC dependence. In particular, A may be the real part or the magnitude of the impedance at a first frequency f1, and B may be, in particular, the imaginary part or the phase at the frequency f1 and / or the real part or the amount at the frequency f2.

reference data

According to the invention, a simultaneous determination of both T and SOC of a lithium-ion cell by measuring at least two impedance values A and B with different T / SOC dependence is possible, as long as the T / SOC dependence of the cell is known.

It is therefore necessary for the method according to the invention to provide impedance reference data for A and B which make it possible to use A or B as a function of T and SOC, ie A _ref (T, SOC) and B _ref (T, SOC), to determine. The reference data may be previously determined by adjusting the cell to a plurality of different, well-defined T and SOC values, for example by thermostating and controlling the amount of charge flow, and the values of A and B for the respective set T-SOC Condition can be measured.

The reference points thus obtained can be stored, for example, in the measuring device or the control unit, which carries out the evaluation, as a look-up table (LUT). The determination of A _ref (T, SOC) and B _ref (T, SOC) can then take place on the basis of the reference points from the LUT by reading or interpolation.

As an alternative, it is also possible to generate the reference data model-based. The procedure is described in the literature, for example in Hu et al, "A comparative study of equivalent circuit models for Li-ion batteries," 2012, Journal of Power Sources, Vol. 198, pp. 359-367 or in Seaman et al, "A survey of mathematics-based equivalent-circuit and electrochemical battery models for hybrid and electric vehicle simulation", 2014, Journal of Power Sources, Vol. 256, pp. 410-423 ,

The methods described there can be used for the method according to the invention. In particular, methods can be used in which the behavior of the cell is modeled by an equivalent circuit diagram, which includes as components at least ohmic resistances R and capacitances C.

Since the reference data required according to the invention need only be predicted for the measurement frequency (s), it may be possible to considerably simplify the models described in the literature by omitting components with significantly longer time constants, as are necessary, for example, for the description of diffusion processes become. For example, a suitable model may include only a series resistor and one or more RC elements. The temperature and SOC dependence of R and C can be represented by equations to be determined empirically, optionally also by analytical approaches, e.g. Arrhenius behavior in the electrolyte conductivity.

The free parameters of the model (ie the various round C values as well as the SOC and T dependency parameters) can be obtained by fitting the model to reference points where the reference points can be obtained in the same way as described for the LUT-based approach. Mixed forms, such as an analytic prediction of T-dependence using an Arrhenius model, in combination with a LUT for SOC dependence, are also possible.

In general, A and B may depend not only on T and SOC, but also on other parameters such as the state of health (SOH) of the cell. Such dependencies preferably also flow into the determination of A _ref (T, SOC) and B _ref (T, SOC) from the reference points. For this purpose, the LUT or the impedance model can be supplemented by further reference points for cells with a defined state of aging in order to use reference data for A and B as a function of T, SOC and SOH, ie, A _ref (T, SOC, SOH) and B _ref (FIG. T, SOC, SOH), to read or calculate.

The aging state (SOH) of the cell, whose temperature and SOC are to be measured by the method according to the invention, is typically estimated by the battery management system routinely from the history and current / voltage behavior of the cell and is therefore known. In addition, the SOH can optionally be determined, for example, by measuring the usable capacity after maximum charge. Thus, the SOH can be used as a previously known constant, and the reference data are again simplified to A _ref (T, SOC, SOH = const) = A _ref '(T, SOC) or B _ref (T, SOC, SOH = const) = B _ref '(T, SOC).

It is also possible, in the meantime, at approximately predetermined intervals, to bring the cell to a defined state of charge and to thermally equilibrate it, and then to determine it for the T-SOC state A and B thus set. The calibration points thus obtained can then likewise be stored in the respective control unit and, if appropriate, used to correct the reference data read or interpolated from the LUT or to update the parameters of the impedance model. In particular, the calibration points may be performed with a known SOH determined by the battery control system to enable a correction with respect to the SOH.

measuring frequency

The excitation frequency in the impedance measurement is typically in the range of 1 Hz to 10 kHz, preferably 50 Hz to 5 kHz, in particular 100 Hz to 1 kHz. The choice of frequency can exert an influence on the precision of the SOC and T determination in the method according to the invention.

Thus, the extent of the SOC dependence of the impedance depends strongly on the frequency. At a frequency of 850 Hz, as in the in 1 and 2 In the case of the measurements shown, Re (Z) falls in the range of 0 to 100% SOC by about 0.1 mΩ and Im (Z) rises in the same range by about 0.05 mΩ. Regarding the T-dependence, Re (Z) decreases in the range of -20 ° C to + 40 ° C by about 0.9 mΩ and Im (Z) decreases by about 0.4 mΩ. As the frequency increases, the SOC dependence generally decreases less than the T-dependence. This means that the isolines deviate less from the parallels to the SOC axis and thus cover a narrower T-range for a given value of Re (Z) or Im (Z). Thus, the accuracy of the T determination increases, and the accuracy of the SOC determination decreases. With regard to an accurate T determination, therefore, higher frequencies are preferred, for example 500 Hz to 10 KHz, in particular 1 kHz to 5 kHz.

Conversely, the SOC dependence and thus the accuracy of the SOC determination increase as the frequency decreases. At low frequencies, for example 500 Hz or less, the SOC can thus be more accurately determined. However, the temperature determination becomes increasingly inaccurate, because although the temperature dependence of the impedance also increases, but the influence of SOC and SOH can not be eliminated, as is the case at the high frequencies.

In a preferred embodiment, multiple frequency measurements may be performed to improve accuracy. From a first measurement at high frequency, for example 1 kHz to 5 kHz, an accurate temperature determination and a rough determination of the SOC can be made. With a second measurement at low frequency, for example 10 to 500 Hz and including the temperature determined in the first measurement, the SOC can then be determined exactly.

Determination of T and SOC

In the method according to the invention, T and SOC are determined as those values at which ΔA = AA _ref (T, SOC) and ΔB = BB _ref (T, SOC) become simultaneously minimal, which in particular implies the case where ΔA and ΔB are simultaneously zero ,

As in the 1-3 illustrates that the set of T-SOC pairs where ΔA = 0 defines an isoline for A, the set of T-SOC pairs ΔB = 0 defines an isoline for B, and a T-SOC pair for which ΔA and ΔB are zero at the same time, corresponds to an intersection of the isolines.

In the simplest case, the method according to the invention only involves the measurement of two impedance values A and B. In this case, the isolines typically only intersect at one point, ie there is a value pair for T and SOC for which ΔA = 0 = ΔB. To improve accuracy, this section of the isolines should be as steep as possible.

Thus, A can be the real part and B the imaginary part of the impedance at a frequency f1 with as much as possible a different course of the isolines, for example in the range from 10 to 1 kHz. 4 outlines the procedure for this case.

Alternatively, A may be the real part or the magnitude of the impedance at a frequency f1 with weak SOC dependency, for example in the range of 1kHz or more, and B may be the real part or the amount in a frequency range f2 with strong SOC dependency, for example 10 to 500 Hz. Preferably, f1 is at least 500 Hz higher than f2.

In this case, the isoline for A differs only slightly from the parallel to the SOC axis, ie the covered temperature range ΔT in the interval from 0 to 100% SOC is low, for example 5 ° C or less. By contrast, B has a steeper angle to the SOC axis (and thus also to the isoline for A) and covers, for example, a ΔT range of 20-30 ° C., as in FIG 1 shown. In this way, even from A alone, the temperature can be estimated with relatively high accuracy, and the relatively steep angle between the isolines then allows a high accuracy in the estimation of the SOC. The relatively low SOC dependence of A has the further advantage in this case that less SOC measurement points are required for the preparation of the reference data for A.

To improve the measurement accuracy, it is also possible to use several B values, for example the imaginary part or the phase at the frequency f1 of the A value as the first B value, as well as the real part or magnitude and / or B values the imaginary part or the phase at one or more further frequencies f2.

When projecting the A and B values into the T-SOC level, more than two isolines are obtained, which in practice no longer have to intersect in one point. In this case, the determination of T and SOC represents a numerical optimization task involving the simultaneous minimization of ΔA and all ΔB. Different quality criteria can also be used for the minimization, for example the ΔA or ΔB with the largest T-dependency and / or the weakest SOC dependency can be most heavily weighted with regard to optimizing the T-value.

For example, the sum of the deviation squares of A and the different B values B ₁ , B ₂ , etc., can be minimized: $$\underset{T.SOC}{\text{min}}\left[a{\left(\text{A}-{\text{A}}_{ref}\left(\text{T}.\text{SOC}\right)\right)}^2+{\text{b}}_1{\left({\text{B}}_1-{\text{<figure-callout id="1" label="B" filenames="DE102017218715A1\_0002.png" state="{{state}}">B</figure-callout>}}_{\mathrm{1,}ref}\left(\text{T}.\text{SOC}\right)\right)}^2+\cdots \right]$$

Here, a, b _1, etc. represent the optional weighting factors.

Apparatus for carrying out the method

• - einen Signalgenerator zur Erzeugung eines Anregungssignals in einer oder mehreren Frequenzen f1 bzw. f2, das an die Zelle angelegt wird;
• - eine Messeinheit zur Messung des durch das Anregungssignal verursachten Antwortsignals;
• - eine Auswertungseinheit zur Bestimmung der Impedanzwerte A und B aus dem im Signalgenerator erzeugten Anregungssignal und dem in der Messeinheit gemessenen Antwortsignals;
• - eine Einheit zur Bereitstellung von Referenzdaten für A und B in Abhängigkeit von SOC und T;
• - ein Einheit, das dafür eingerichtet ist, aus den in der Auswertungseinheit bestimmten Impedanzwerten A und B und den bereitgestellten Referenzdaten SOC und T zu bestimmen.

An apparatus for carrying out the method according to the invention comprises at least the following constituents:

• a signal generator for generating an excitation signal in one or more frequencies f1 or f2, which is applied to the cell;
• a measuring unit for measuring the response signal caused by the excitation signal;
• an evaluation unit for determining the impedance values A and B from the excitation signal generated in the signal generator and the response signal measured in the measuring unit;
• a unit for providing reference data for A and B as a function of SOC and T;
• a unit adapted to determine from the impedance values A and B determined in the evaluation unit and the reference data SOC and T provided.

Preferably, this device is part of a battery system, in particular a battery system for an electrically operated vehicle.

In such a battery system, the cells are typically connected in series and / or parallel to battery packs and each connected to a cell monitoring unit (CSC) which monitors at least the cell voltage and also controls the charge balancing. In this case, each individual cell can be provided with a cell monitoring unit, or a plurality of cells can be connected to a cell monitoring unit. This can have several input channels for voltage measurement in order to be able to monitor the cells connected to it at the same time, or the monitoring can take place via a multiplex method. The whole of the cells and cell monitoring units is in turn monitored by a battery management unit (BMS).

An impedance ^{measurement basically requires the application of} an excitation signal (typically an AC signal I (t) = I ₀ * e ^iωt ), the recording of the response signal (typically as Voltage signal U (t) = Uo * e ^{iωt + φ} and the calculation of the complex impedance Z = U (t) / I (t) = (Uo / Io) * e ^iφ .

In the device according to the invention, the signal generator for generating the excitation signal can be integrated in the cell monitoring unit, which, for example, impresses an alternating current signal on the balancing current. Alternatively, the excitation signal can be globally modulated onto the outside of it. The measurement of the response signal is expediently carried out by the cell monitoring unit.

The calculation of the impedance from the current and voltage signal, the comparison with the reference data and the determination of T and SOC may also be done in the cell monitoring unit, or they may be performed partially or completely by another controller, for example the battery management unit.

The device further comprises a unit for providing the reference data. This can include a data memory in which the values for A and B are stored as a function of SOC and T as a table (LUT), and from which the required reference data can be read out or interpolated. Alternatively, the unit may comprise a computer suitably programmed to calculate the reference data based on an impedance model having previously known parameters. The unit for providing the reference data may also be integrated into the cell monitoring unit. Preferably, however, this function is taken over by the battery management unit.

As a unit for determining T and SOC, both the cell monitoring unit and the battery management unit can be used, the latter being preferred.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2009/0096459 [0004]
• DE 102009038663 [0005]
• DE 102013103921 [0007]
• EP 2667166 A2
• US 2013/0264999 [0009]

Cited non-patent literature

• Hu et al, "A comparative study of equivalent circuit models for Li-ion batteries," 2012, Journal of Power Sources, Vol. 198, pp. 359-367. [0027]
• Seaman et al, "A survey of mathematics-based equivalent-circuit and electrochemical battery models for hybrid and electric vehicle simulation", 2014, Journal of Power Sources, Vol. 256, pp. 410-423 [0027]

Claims (11)

A method of determining temperature T and state of charge SOC of a lithium-ion cell, comprising: (a) measuring the real part or magnitude of the impedance Z of the cell at a frequency f1 to obtain a first impedance value A; (b) measuring at least one further impedance value B selected from: - imaginary part or phase of the impedance Z at the frequency f1 - real part or magnitude of the impedance Z at another frequency f2 chosen such that the T-dependence and / or distinguishes the SOC dependence of the impedance from that at f1; Imaginary part or phase of the impedance Z at the frequency f2; (c) providing impedance reference data for the impedance values measured in (a) and (b) as a function of T and SOC, A _ref (T, SOC) and B _ref (T, SOC), the impedance reference data being from previously known cell-specific calibration data are read, interpolated or calculated; (d) Determination of T and SOC from the impedance values A and B measured in a and b by comparison with the reference data, where T and SOC are those where ΔA = AA _ref (T, SOC) and ΔB = BB _ref (T , SOC) at the same time become minimal. Method according to Claim 1 , characterized in that T and SOC are obtainable by the following steps: - determining those T-SOC value pairs from the reference data which satisfy A _ref (T, SOC) = A to obtain a T-SOC isoline for A ; - determining those T-SOC value pairs from the reference data which satisfy B _ref (T, SOC) = B to obtain a T-SOC isoline for B; - Determination of T and SOC as the intersection of the isolines. Method according to Claim 1 or 2 where f1 is in the range of 10 Hz to 1 kHz. Method according to Claim 1 wherein B comprises at least the real part or magnitude of the impedance at a frequency f2 in the range of 10 to 500 Hz, and f1 is at least 500 Hz higher than f2. Method according to at least one of Claims 1 to 4 wherein the reference data are read from a predetermined look-up table (LUT) and possibly interpolated, wherein the table is set by prior adjustment of the cell to known T or SOC states and measurement of the impedance values for the respectively set state is obtained. Method according to at least one of Claims 1 to 4 wherein the reference data is calculated based on a predetermined parameterized impedance model of the cell, wherein the parameters of the model are obtainable by fitting to reference points by adjusting the cell to known T and SOC states and measuring the impedance values for the respectively set state to be obtained. Method according to at least one of Claims 1 to 6 wherein the cell is set at predetermined time intervals to a defined temperature and a defined SOC and a calibration measurement of A and B is performed to correct the reference data. Method according to at least one of Claims 1 to 7 in which, in addition to the dependence of T and SOC, the reference data also includes the dependence of A and B on the state of aging of the cell SOH, and the method further comprises: - providing cell aging status (SOH) data obtained by a per se known method the history of the cell, the current / voltage behavior and / or a measurement of the remaining useful capacity are determined after maximum charging; Determination of T and SOC from the measured impedance values by comparison with the reference data for A and B at the respectively determined SOH value. Apparatus for carrying out the method according to at least one of Claims 1 to 8th comprising: a signal generator for generating an excitation signal at one or more frequencies f1 and f2, respectively, applied to the cell; a measuring unit for measuring the response signal caused by the excitation signal; an evaluation unit for determining the impedance values A and B from the excitation signal generated in the signal generator and the response signal measured in the measuring unit; a unit for providing reference data for A and B as a function of SOC and T; a unit which is set up from the impedance values A and B determined in the evaluation unit and the provided reference data according to the method according to at least one of Claims 1 to 8th Determine SOC and T. Device according to Claim 9 in which the unit for providing the reference data comprises a data memory in which the previously known reference data are stored as a table. Device after Claim 9 wherein the reference data providing unit comprises a computer adapted to store the reference data calculated on the basis of an impedance model with previously known parameters.

Images