EP3356835A1

EP3356835A1 - Method for determining a real component of a complex internal resistance of a battery

Info

Publication number: EP3356835A1
Application number: EP16774935.7A
Authority: EP
Inventors: Hans-Michael Graf; Thomas Klement; Martin Schramme
Original assignee: Continental Automotive GmbH
Current assignee: Continental Automotive Technologies GmbH
Priority date: 2015-09-29
Filing date: 2016-09-29
Publication date: 2018-08-08
Also published as: CN108291943A; CN108291943B; WO2017055428A1; DE102015218797A1

Abstract

The invention relates to a method for determining a real component of a complex internal resistance of a battery, wherein associated battery voltage values and battery current values are measured at a plurality of measuring time points, said values being produced corresponding to a number of connected consumers, and wherein a frequency-independent real component is determined using a calculation rule based on said battery voltage values and battery current values.

Description

Method for determining a real part of a complex internal resistance of a battery

The invention relates to a method for determining a relative portion of a complex internal resistance of a battery. In particular, this can be understood as the determination by a battery sensor (IBS) connected to the battery.

An internal resistance of a battery typically describes the magnitude of voltage changes in relation to the current changes applied to the battery, for example according to Ohm's Law. For the diagnosis of a lead battery, the real part of the internal resistance of the battery above an excitation frequency of about 100 Hz is particularly interesting. At significantly higher frequencies, inductive components contribute to the complex resistance of the battery, and at significantly lower frequencies, capacitive components.

In the prior art, it is known that to determine the resistance at a frequency, a signal of known frequency is applied to the battery or applied to a broadband signal spectrum and then the behavior is filtered out in the interesting frequency range. In a motor vehicle, in particular, offers the latter method. In the laboratory, however, the former method is usually used.

An intelligent battery sensor (IBS) offers the possibility of measuring current and voltage at a frequency of at least 1000 Hz and by evaluating both signals to infer the real part of the resistance. In the intelligent battery sensor, the battery current is usually ge ^¬ measured using a shunt resistor. The voltage signal at the shunt resistor is amplified and amplified via a fast, time-discrete analog-digital (AD) converter converted to digital values. The battery voltage is fed via a voltage divider into an AD converter.

So far, the calculation of an internal resistance of a battery has typically been performed as follows. An apparatus consisting of a microcontroller, an analog amplifier, an AD converter and a shunt resistor provides measurement signals of current and voltage, which meet the phy ^¬ sikalischen conditions to perform a Wider- standsberechnung. Thus, current and voltage are available as time-discrete value pairs. The current and voltage values that are available result from the activity of the devices connected to the vehicle electrical system such as battery, control units, generator or headlights. Thus, the battery sensor typically does not supply and evaluate an active measurement signal to the battery.

The measured current and voltage values undergo filtering, are differentiated and the quotient is formed. This is done according to Ohm's law R = dU / dI. A suitable downstream averaging and normalization can then lead to the final value of the internal resistance. In this determination method, however, the real part of the internal resistance is not determined, but rather the absolute value of the complex impedance Z, as shown schematically in FIG. 1, for example. In addition, the exciting frequency is given by the electrical system, which makes it variable and unpredictable. This leads to strong fluctuations. Alternatively, the determination of the internal resistance, in particular with laboratory means, can usually be carried out with a special measuring device which determines the complex impedance Z (f) at a frequency of, for example, 1 kHz. For this purpose, a small sinusoidal current, usually with 1000 Hz, of the battery becomes active impressed. The voltage change caused by this current on the battery is measured and the real and imaginary part of the impedance Z are determined by means of the resulting phase shift.

The resulting real resistance R is then the real resistance of the battery. The problem for an application in a battery sensor is the comparatively high effort to actively impart a current, so that this method is predominantly found in laboratory measuring instruments.

It is therefore an object of the invention to provide a method for determining a real part of a complex internal resistance of a battery, which actually provides the real part and is easier to use.

This is achieved according to the invention by a method according to claim 1. Advantageous embodiments can be taken, for example, the dependent claims. The content of the claims is incorporated by reference herein express the content of Be ^¬ sensitive.

The invention relates to a method for determining a real part of a complex internal resistance of a battery, comprising the following steps:

Measuring, at a plurality of measurement times (t ±), in each case an associated battery voltage ^value (U) and an associated battery current ^value (I), which result ent ^¬ accordingly a number of connected consumers, and

Determining a frequency-independent real part using a calculation rule based on the battery voltage values (U) and the battery current values (I) The calculation of the real part of the internal resistance is based on a differential equation based on the battery model. For this purpose, the differential equations associated with the battery model are determined. The calculated differential equations are converted into their time-discrete counterpart, the difference equations, since the measured values present for the calculation of the internal resistance are, in particular, time-discrete.

In a further step, these difference equations or differential equations are resolved according to the real part of the complex internal resistance of the battery to be determined.

By means of the method according to the invention, in contrast to the first method known from the prior art as described above, an accurate determination of the real part and not only the amount of the impedance is possible. In contrast to the second method described above with excitation, the current is measured according to connected consumers and not according to excitation. This can be dispensed with a suggestion.

The inventors of the present application have recognized that calculation rules exist which allow such a procedure. From the prior art, however, no such calculation rules are known, but it was in the prior art as not possible to measure separately without stimulation a real part.

It should be understood that instead of a frequency-independent real part can also be simply spoken of a real part. To connected devices can be, for example, components that are typically present in a vehicle, in particular electronic Steuerungsvor ^¬ directions, lights, alternators, other components of a vehicle lighting, ignition or comfort items such as electric windows or motors of an electric seat adjustment.

In particular, the measurement is performed without exciting the battery or imparting an excitation signal to it. This can be understood in particular to mean that there are no components which specifically impose a current or a voltage of the battery for the purpose of measurement. On the provision of such components, which are typically expensive, can thus be dispensed with, which significantly improves the usability of the method, especially in vehicles for the mass market.

The rule may, in particular based on a He ^¬ equivalent circuit diagram of the battery. Such a replacement circuit diagram will be described below and also with reference to the drawing at ^¬ , so that reference can be made to the comments below.

At least the following parameters can advantageously be included in the calculation rule:

Battery model inductance (L),

Battery model capacity (C _x ),

parasitic resistance and battery capacity (R _x ). It has been found that calculation rules in which these parameters are incorporated make it possible to determine the real part in a particularly advantageous manner, that is to say in particular simply and accurately. Such model parameters L, R _x and C _x can in particular be functions of a state of charge (SOC), a temperature and / or a battery capacity.

Advantageously, at least the battery voltage values (U), the battery current values (I), first and second derivatives of the battery voltage values (U) can be included in the calculation specification first, second and third time derivatives of the battery ^¬ current values (I) received. Calculation rules with such values have proven to be advantageous for typical applications. Exemplary calculation rules are given below.

According to one embodiment, the calculation rule may be given by the following formula:

^ U (t) -C _x -R _{x +} ± U (t) + l (t) ^■ L - ^ I (t) -R _x -C _x -L - ± I (t) * R _x

^ _ dt dt dt dt dt

-II ((tt)) ++ - _r I (t) -C _x -R _x

dt dt

This procedure is particularly advantageous if it can be assumed that the open terminal voltage (OCV = Open Clamp Voltage) of the battery is constant. This procedure represents a particularly simple embodiment, in particular in comparison to the embodiments described below, in which this condition does not apply.

According to one embodiment, the calculation rule can specify that the real part is to be determined as the slope of a straight line which approximates points in a two-dimensional coordinate system having an x-axis and a y-axis, wherein the calculation ^rule plots applied terms on the x-axis and indicating on the y-axis. By means of such a calculation rule, in particular in the case in which OCV is not constant or can not be assumed to have a constancy of OCV, the real part can be determined. Examples of this are given below. The calculation rule can specify in particular that the term dt dt is plotted on the x-axis and the term is plotted on the y-axis dt dt dt dt dt is applied. Such a procedure has proved particularly advantageous in the case in which OCV can not be assumed to be constant.

By minimizing the distances between a plurality of measurement points, this results in a regression line with the following Straight ^¬ engleichung:

v _V _V n rn ^ d ² OCV dOCV

Y = aX + b = R - X + R _X C _X - +

dt dt

The slope a thus corresponds to the real resistance R.

According to an alternative calculation rule, this can also indicate, for example, that the term dt dt is plotted on the x axis and the term dt dt dt dt is plotted on the y axis.

The calculation rule may indicate in particular that at least two different regression methods are used for loading ^¬ humor of a respective real part, wherein a distance between the respectively determined real parts is a measure of the quality of the calculation. Thus, the Zuver ^¬ permeability can be increased in the determination of the real part. However, it should be understood that basically only one regression method can be used. ₀

O

Regarding the regression method to be used for determining a straight line slope, known methods can be used. Examples of this are described below.

In particular, the calculation rule can specify that the real part is calculated according to the following formula:

Σ ^χ _{ ^{■ γ} ι Zfo + *,) ^■ (y _t + Ay _t ) Σ & + *,) ^■ (Sx _t + Ay _t )

__i __i

Σ (ϋχ, χ, + χ, (RAx, + Ay,) + Αγ, Αχ,) in which

Xi are the values of the term plotted on the x-axis at time t ±,

yi are the values of the y-axis term at time t ±, and

- Δχι and Δγι parameters are.

By means of the formula just described, in particular the noise can be taken into account, mean value differences never becoming zero even under ideal conditions. To improve the regression result, therefore, it is possible, for example, to correct the sums caused by the noise of current and voltage as just described.

The calculation rule can also specify that the real part is calculated at each measurement time as a quotient of a first term and a second term. Again, suitable formulas have been found which are suitable for determining a real part. For example, the calculation ^{rule may} specify that the first term dt dt is and the second term is.

Thus, the computationally expensive regression can be dispensed with and the real part can instead be determined by a simple division to be performed. This can be at ^¬ play as be advantageous to be as economical with computer capacity, for example, the cost of a corresponding calculation module not to push too high. The battery may advantageously be a lead-acid battery. Such batteries have proven themselves in particular for use in motor vehicles, and it has been shown that the inventive method is applicable before ^¬ geous, especially for such batteries. It should be understood, however, that under a battery, typically a rechargeable battery is meant here, which may be referred to for example as Ak ^¬ kumulator. It should be noted in this regard that for batteries used in motor vehicles, the term battery, especially car battery, has prevailed.

The battery voltage values (U) and / or the battery current values (I) may be used filtered or unfiltered. Filtering can improve the accuracy of the calculation, while eliminating filtering can reduce the need for computational power. According to a further aspect of the invention, a control device is set up to carry out a method according to one of the preceding claims.

In a development of the specified control device, the specified device has a memory and a processor. In this case, the specified method is stored in the form of a Compu ^¬ terprogramms in the memory and the processor is provided for performing the method when the computer program from the memory is loaded into the processor.

According to a further aspect of the invention, a computer program comprises program code means for performing all the steps of one of the specified methods when the computer program is executed on a computer or one of the specified devices.

According to a further aspect of the invention a computer program product comprises a program code which is stored on a data carrier and the compu ^¬ terlesbaren, when executed on a data processing device, carries out one of the methods specified.

According to another aspect of the invention, a battery sensor comprises a specified control device.

According to another aspect of the invention, a vehicle includes a specified battery sensor.

The invention further relates to a battery arrangement with a battery, in particular a lead-acid battery and such a device. Moreover, the invention relates to a computer-readable non-volatile storage medium containing program code, in the execution of which a processor performs the method according to the invention. With respect to the driving Ver ^¬ thereby embodiments and variants described in each case on all herein may be resorted to. The invention will be described in more detail below with reference to the accompanying drawings. Showing:

1 shows a typical course of resistance of a lead-acid battery at different frequencies,

Fig. 2: an electrical equivalent circuit diagram of a lead-acid battery.

The invention is based on an electrical equivalent circuit diagram of the battery, which is parameterized such that it maps the nen ^¬ resistance of the battery to be measured. Such an equivalent circuit diagram is reproduced in FIG. 2, where L denotes an inductance, R the internal resistance, C _x a capacitance, R _x a capacitance C _x parasitic resistance, and OCV an open terminal voltage. The open terminal voltage OCV may in particular depend on parameters such as a temperature T or a state of charge SOC.

The model can also be extended by additional elements compared to Figure 2: resistors R, capacitances C, inductors L, nonlinear elements, Warburg element, etc., It can also be parameterized depending on the battery state (state of charge, temperature, etc.).

It can be assumed typically that over the service life of the battery Le ^¬ particular the real part of the internal resistance R increases. All other parts of the spare ^¬ diagram can be as nearly constant is accepted ^¬ typically.

The calculation of the real part of the internal resistance then takes place on the basis of a differential equation based on the battery model. For this purpose, the differential equations associated with the battery model are determined, and in a further step these equations are resolved according to the value R to be determined. It is to be understood that hereinafter, instead of the already guided ^¬ designations U and I for the measured voltage and the measured current _batt also the designations U and I are used _batt.

A derivation of a possible differential equation

for example, as follows: u _Rx = u, ^dU R * ₌ ¹ Cx

cx

c

u RCx U _RCX Q dU _RCx _ _R r ^d U _RCx

batt batt U RCx ^R Jbatt

R dt dt

U _bat = OCV + u _RCx = u _batt - This can be solved for example by means of derivative by time as follows:

OCV + R - l _batt + L ie batt

dU x batt - U b, att

RCx dt ocv batt L dl batt R ^~ xJ batt U b, dt _r R CRRC dt R _r CRC dl batt

dt

The value R represents the real internal resistance of the battery. If OCV were constant, then the real resistance at any time by means of the following simplifying ^¬ fanned equation could be calculated:

The derivatives of current and voltage after the time (d / dt) can be calculated, for example, in a microcontroller (yC) by differentiation.

The value R is then independent of frequency.

Often, however, the assumption that OCV is constant will not be sustainable. In this case, the real resistance can be determined, for example, by a least square analysis (regression line). According to one embodiment, the term X determined from measured values is determined on a virtual x-axis

_ dl batt _ | _ _Q Ibatt

dt ^x dt ²

applied. The virtual y-axis is used to plot the calculated term Y

γ _ dU batt g _Q d U batt _ g dl b _a tt _ I _batt "^ d I _batt

dt ^x dt ^{2 x} dt dt ^{2 x} dt

By minimizing the distances between a plurality of measurement points results in a regression line with the following Straight ^¬ engleichung: dt dt

The slope a thus corresponds to the real resistance R

The determination of the real part of the internal resistance of a battery, in particular a lead-acid battery, can be More generally, for example, by applying a method with the following steps:

1) Creation of an electrical equivalent circuit

2) Determination of the differential equation of the equivalent circuit

3) conversion of the differential equations in a linear form which aufweiset as a proportionality constant the internal resistance ^¬

4) numerical calculation of all necessary differential quotients a) from the measuring signals of current and voltage and b) from battery specific parameters like L, R _x , C _x , temperature, state of charge, etc

5) numerical determination of the real part of the internal resistance with the aid of the determined differential quotient.

It should be understood that this method is a separate aspect of the invention and may be combined in any manner and in any sub-combination with other aspects disclosed in this application.

An advantage which results from the described solution, is compensated by a phase and frequency behavior of the battery value of internal resistance, whose value is stable with respect to frequency fluctuations ^¬ the vehicle electrical system.

Furthermore, the calculation of the differential quotients even with high sampling rate of a low-cost microcontroller to accomplish.

Required for such a result active launch of measuring signals is omitted in the battery, which also reduces the ap ^¬ para tive effort. It should be mentioned that according to an alternative embodiment it is also possible to evaluate the vehicle electrical system according to frequency and phase, in particular by means of a Fourier transformation. However, this requires a high computing power and thus the use of a DSP (Digital Signal Processor) and the associated higher costs.

By means of the Fourier transformation, in particular a fast Fourier transformation (FFT), knowledge of the frequency and phase of the signals can be minimized, by means of a suitable ^subsequent filtering, whose influence is minimized.

According to one embodiment, the resistance at each new measurement can be calculated from R (t) = Y (t) / X (t). According to another embodiment, the resistance may be determined by regression analysis of a plurality of values of Xi and Y ±.

Instead of the simple derivation after the time

For example, derivatives of higher order time may also be used, for example, as follows:

For finding the equalization line between Xi and Y, for example, the squared distances of the Yi values of calculated Y values can be used: i

For example, the quadratic distances of the Xi values of calculated X values can be used to find the balance line between Xi and Y ±:

For finding the best-fit line between Xi and Y ±, for example, the quadratic distances of the value pairs Xi Yi can also be minimized from the calculated equalization line:

The difference between the resistance R from different calculation methods can be used as a measure of the quality or the error of the resistance.

The calculation of the regression line can be implemented numerically for one of the embodiments described above as follows:

^• (x, Y, -Rx x, - (r-Rx x,) χχ

YX Σ i

To determine the best-fit line, offset B can be set to 0, which simplifies the calculation (here, for the quadratic Yi values of calculated Y values):

-Y (Y, -RX,) ² = 0 => = RX

'^l>

The calculation of the time derivative can be determined, for example, by subtraction:

smell _| At \ _U (t + At) -U (t)

dt 2) At

U (t + At) -U (t) U (t) -U (t-At)

d ^2JJ l _t \ At At _U (t + At) -2-U (t) + U (t-At)

dt ² Δί At ²

Due to the well-known noise of current and voltage, the above mean value differences Xavg * Xavg-Xi * Xi, Yavg * Yavg-Yi * Yi or Xavg * Yavg-Xi * Yi never become 0 even under ideal conditions become. This allows the following formula for calculating the resistance:

Σ ^χ _{ ^{■ γ} ι Zfo + *,) ^■ (y _t + Ay _t ) + *,) ^■ (Sx _t + Ay _t )

__i __i

Σ ^x i · ^x i Σ ( ^χ ι ^{+ Αχ} ι - ( ^χ ι ^{+ Αχ} ι Σ ( ^χ ι ^{+ Αχ} ι - ( ^χ ι +)

i i i

Σ (Rx _iXi + x _t (RAx, + Ay,) + Δ ^ .Δχ) Therefore, to improve the regression result, a correction of the sums caused by noise current and voltage can be made in ^¬ play.

Instead of the equivalent circuit diagram shown in FIG. 2, alternative equivalent circuit diagrams can be taken as a basis. In particular, individual components of the equivalent circuit diagram in Figure 2 can also be omitted.

The input values of current and voltage can be both filtered and unfiltered in the regression calculation.

Mentioned steps of the method according to the invention can be carried out in the order given. However, they can also be executed in a different order. The method according to the invention can be carried out in one of its embodiments, for example with a specific combination of steps, in such a way that no further steps are carried out. However, in principle also further steps can be carried out, even those which are not mentioned.

The claims belonging to the application do not constitute a waiver of the achievement of further protection. If, in the course of the procedure, it turns out that a feature or a group of features is not absolutely necessary, it is already desired on the applicant side to formulate at least one independent claim which no longer has the feature or the group of features. This may, for example, be a subcombination of a claim present at the filing date or a subcombination of a claim limited by further features of a claim present at the filing date. Such newly formulated claims or feature combinations are to be understood as covered by the disclosure of this application.

It should also be noted that embodiments, features and variants of the invention, which are described in the various embodiments or embodiments and / or shown in the figures, can be combined with each other as desired. Single or multiple features are arbitrarily interchangeable. Resulting combinations of features are to be understood as covered by the disclosure of this application.

Recoveries in dependent claims are not to be understood as a waiver of obtaining independent, objective protection for the features of the dependent claims. These features can also be combined as desired with other features.

Features that are disclosed only in the specification or features that are disclosed in the specification or in a claim only in conjunction with other features may, in principle, be of independent significance to the invention. They can therefore also be included individually in claims to distinguish them from the prior art.

Claims

claims

A method of determining a real part of a complex internal resistance of a battery, comprising the steps of:

Determining a frequency-independent real part using a calculation rule based on the battery voltage values (U) and the battery current values (I)

2. The method according to claim 1,

wherein the measuring is performed without exciting the battery or imparting an excitation signal to it.

3. The method according to any one of the preceding claims,

wherein the calculation rule is based on an equivalent circuit diagram of the battery.

4. The method according to any one of the preceding claims,

at least the following parameters are included in the calculation rule:

Battery model inductance (L),

Battery model capacity (C _x ),

parasitic resistance of the battery capacity (R _x ).

5. The method according to any one of the preceding claims,

wherein in the calculation rule at least the Batte ^¬ riespannungswerte (U), the battery current values (I), first and second time derivatives of the battery voltage values received (U) and first, second and third time from ^¬ lines of the battery current values (I).

Method according to claims 4 and 5,

where the calculation rule is given by the following formula:

7. The method according to any one of the preceding claims,

wherein the calculation rule specifies that the real part is to be determined as the slope of a straight line which approximates points in a two-dimensional coordinate system having an X-axis and a Y-axis, wherein the calculation ^rule plots applied terms on the X-axis and on the Y-axis Y-axis indicates.

8. Process according to claims 4, 5 and 7,

where the calculation rule specifies that on the X-axis the term dt dt

is plotted and on the y-axis the term

_Y Lt _{+ C} , - ^d d ^ t - - ^d dL-L

t - ^d -R, C, L

d dt - ^d dt

is applied. 9. Method according to claims 4, 5 and 7,

where the calculation rule indicates that on the x-axis the term

X = - + R _X C _X - ^ ^~

dt dt is plotted on the Y-axis and the dt dt dt dt dt

is applied.

10. The method according to any one of claims 7 to 9,

wherein the calculation rule specifies that at least two different regression methods are used to determine a respective real part, wherein a distance between the real parts represents a measure of the quality of the calculation.

11. The method according to any one of claims 7 to 10,

where the calculation rule specifies that the real part is calculated according to the following formula:

R

i i

X (Rx _iXi + _Xi (RAx, + Ay,) + Δ ^, Δ,)

in which

Xi are the values of the X-axis term at time ti,

yi are values of the term plotted on the y-axis at time ti, and

Dxi as well as Dyi parameters are.

Method according to one of the preceding claims, wherein the calculation rule specifies that the real part is calculated at each measurement time as a quotient of a first term and a second term.

13. The method according to claim 12,

wherein the calculation rule indicates that the first term dt dt

is and the second term

is.

14. The method according to any one of the preceding claims,

the battery is a lead-acid battery.

15. The method according to any one of the preceding claims,

wherein the filtered battery voltage values (U) and / or the Bat ^¬ teriestromwerte (I) or unfiltered used.

16. A control device which is adapted to carry out a method according to one of the preceding claims.