EP2052271A1

EP2052271A1 - Method for determining the battery capacity using capacity-dependent parameters

Info

Publication number: EP2052271A1
Application number: EP07730093A
Authority: EP
Inventors: Eberhard Schoch; Burkhard Iske; Michael Merkle
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2006-08-07
Filing date: 2007-06-12
Publication date: 2009-04-29
Also published as: JP2010500539A; KR20090045227A; DE102006036784A1; CN101501518A; US20100066377A1; WO2008017530A1

Abstract

The invention relates to a method for determining a battery size (Qe, Uc0min), in particular the capacity (Qe) of the battery (3), with the aid of a state variable and parameter estimator (1) which computes the state variables (Z) and parameters (P) of a mathematical energy-storage model from operating variables (UBatt, IBatt, TBatt) of the battery (3). The capacity (Qe) of the battery (3) can be determined very accurately during normal operation of the battery if it is calculated as a function of at least one capacity-dependent parameter (RK025,vgr25).

Description

description

title

Method for determining the battery capacity based on capacity-dependent parameters

State of the art

The invention relates to a method for determining a battery size, in particular the capacity, of an energy store according to the preamble of patent claim 1, as well as a corresponding device according to the preamble of patent claim 11.

In motor vehicle electrical systems, the electrical consumers are usually supplied by a battery and a generator with electrical power. During driving, an energy and consumer management is usually performed in which individual consumers are automatically switched on or off as needed, for. B. to respond to supply bottlenecks or to perform certain functions. In the context of this energy and consumer management, the knowledge of the battery condition is essential.

To estimate the state of the battery, it is known to use mathematical battery models that describe the electrical and physical properties of the energy store. Using a mathematical battery model, for example, the performance (SOF), the state of charge (SOC) or the capacity or removable charge (Qe) can be estimated.

The known from the prior art battery models include a number of state variables and parameters that are constantly adapted to the current state of the battery in the active electrical system operation. Certain parameters of the Battery models, such as the minimum open circuit discharge voltage (U _c omiπ), however, can only be accurately determined at very low battery states, typically below about 50% of the removable charge. Such deep discharges occur in the vehicle electrical system, however, only very rarely and are also prevented by the energy management of the vehicle to the

Do not endanger the starting ability of the vehicle and to keep the battery aging due to excessive cyclization as low as possible.

Disclosure of the invention

It is therefore an object of the present invention to provide a method for determining a battery size, in particular the capacity of the battery, by means of which the desired battery size can also be determined outside a deep discharge state, ie in particular during normal operation of the battery with high accuracy. The method should moreover be able to be carried out in the case of the electrical excitations present in normal on-board electrical system operation, and in particular of any additional excitations, as described e.g. occur at engine start require.

This object is achieved according to the invention by the features specified in the patent claim 1 and in claim 11. Further embodiments of the invention are the subject of dependent claims.

An essential aspect of the invention is to calculate the desired battery size as a function of at least one capacity-dependent parameter. The invention is based on the recognition that change different battery parameters in the course of operation of the battery compared to the new state and thus the parameters or their change is a measure of the battery condition, especially the capacity (loss capacity or available capacity) of the battery. The sought battery size can thus be determined very easily and accurately taking into account the at least one capacity-dependent parameter. In addition, apart from the excitations usually present in on-board network operation, no additional suggestions of the battery, such as actively triggered charging or discharging current pulses, are required in order to carry out the calculation. According to a preferred embodiment of the invention, the sought-after battery size is calculated as a function of a minimum rest voltage, which in turn is a function of at least one capacitance-dependent parameter.

In particular, a parameter (eg RK025) of the acid diffusion resistance (R _K ) of the battery and / or a parameter (eg Vgr25) of the passage resistance (Rd _P ) between electrolyte and an electrode of the battery can be used as capacity-dependent parameters , The mentioned parameters have the particular advantage that they change relatively strongly with increasing capacity loss, ie are relatively sensitive, and can be accurately identified in vehicle electrical system operation without additional active excitation.

The sought-after battery size is preferably calculated as a function of the deviation of one or more capacity-dependent parameters from a reference value, in particular an initial value in the new state of the battery. Of the

Degree of deviation from the reference value is a measure of the loss capacity of the battery.

According to a preferred embodiment of the invention, the at least one capacity-dependent parameter is weighted with a predetermined factor, which preferably depends on the error variance with which the parameter was determined. Some types of state quantity and parameter estimators output not only the individual quantities or parameters, but also the error variance by means of which the state variable or the parameter was estimated. This value can be used for the weighting of the capacity-dependent parameter.

The desired battery size, such as a capacity of the battery, is preferably also calculated as a function of the maximum rest voltage of the fully charged battery. The maximum rest voltage is preferably learned by means of an adaptation algorithm at high states of charge.

The calculation is preferably carried out by means of an extended Kalman filter.

Brief description of the drawings The invention will now be described by way of example with reference to the accompanying drawings. Show it:

1 shows a device for calculating a battery state variable, in particular the removable from the battery charge. and

Fig. 2 is an equivalent circuit diagram for a lead-acid battery

Embodiments of the invention

Fig. 1 shows a device for determining a battery size, such as the removable from a battery charge Qe (capacity). The device essentially comprises a state variable and parameter estimator 1, as well as a charge predictor 2 (estimator), in which a mathematical energy storage model is stored.

The state quantity and parameter estimator 1 calculates from the actual operating quantities of the battery 4, namely the battery voltage Uβatt, the battery current I Batt and the battery temperature Tβatt, state variables Z and / or parameter P, on the basis of which the charge predictor 2 searches for the desired battery state variable Qe, or other sizes such as the state of charge SOC or the capacity SOF of the battery 4 is calculated. The battery 4 in the following example is a lead-acid battery.

In particular internal voltages U, which result from the equivalent circuit diagram of the battery shown in FIG. 2, apply here as state variables Z. In particular, the parameters mentioned are elements of the equivalent circuit, such as Resistors R and capacitances C, or to various values that occur in the functions of the mathematical battery model.

The calculation of the battery capacity Qe is based on the current state of the energy storage. The mathematical models stored in the charge predictor 2 are therefore initially initialized to the current operating state of the energy store 4. For this purpose, the state quantity and parameter estimator 1 supplies the corresponding initial values. When State variable and parameter estimators can be used, for example, a known Kalman filter. In the course of battery operation, the state variables Z and parameter P are constantly adapted to the current state and thus the functions of the battery model are adapted.

2 shows an equivalent circuit diagram of a lead-acid battery 4. The individual quantities are as follows:

Operating variables:

Ißatt battery current Ußatt terminal voltage of the battery

Tßatt acid temperature

State variables:

Uco open-circuit voltage U _k concentration polarization U _D p _{penetration polarization of} the positive electrode U _{0n penetration polarization of} the negative electrode

Parameter:

Ri (Uco, Uk, Tßatt, RiO25, UcOmiπ, UcOmax) Ohmic internal resistance, depending on the rest voltage Uco, the concentration polarization U _{k of} the acid temperature T _Ba tt the to 25 ° C u. Full charge related internal resistance RiO25 and the minimum rest voltage Ucomiπ at discharge and the maximum rest voltage Ucomax at full charge,

C ₀ acid capacity

Rk (Uco, Tßatt, RkO25, UcOmax) acid diffusion resistance, depending on the quiescent voltage Uco, the positive electrode penetration polarization, the acid temperature T _Ba tt, the to 25 ° C u. Full charge related acid diffusion resistance RkO25 and the maximum rest voltage Ucomax at full charge, C _k capacity of acid diffusion,

RD _P (UCO, U _Dp , Tßatt, ID _P , v _gr 25, Ucomax) Passage resistance between positive

Electrode and electrolyte, depending on the rest voltage Uco, the penetration polarization of the positive electrode UD _P , the acid temperature T _Ba tt, the passage current of the pos. Electrode 1 _Dp , the saturation _{voltage of the penetration polarization} v _gr25 , referred to 25 ° C., and the maximum rest voltage Uco _max

Full charge, C _DP double layer capacity between positive

Electrode and electrolyte, R _DΠ (U _DΠ Tßatt, tan) Resistance between negative electrode and electrolyte, depending on the penetration polarization of the negative electrode UD _Π , the acid temperature T _Ba tt and the flow rate of the negative electrode l _Dp C _DΠ Double layer capacitance between negative electrode and Electrolyte.

The individual equivalent circuit sizes are due to various physical effects of the battery 3, which are known in the art from the relevant literature.

The following function can _κ for the acid diffusion resistance R, for example, are applied:

RK = f (Ucθ, Tßatt, RkO25, Ucomax) Here, R _k o25 is the diffusion resistance of the acid at 25 ° C and fully charged battery. RkO25 is a capacity-dependent parameter.

For example, the following function can be used for the passage resistance R _DP between the electrolyte and the positive electrode of the lead-acid battery 3:

Rüp = f (U _c O, UD _P , Tßatt, lDp, Vgr25, Ucθmax)

Vgr25 is the polarization voltage of the positive electrode at high discharge currents and 25 ° C. vgr25 is a capacity-dependent parameter.

For other state variables (eg U _DP , U _DΠ , U _K , etc.) and parameters (eg R _0n , Co, Ri, etc.) the charge predictor 2 comprises correspondingly different mathematical approaches.

Basically, the following relationship applies to the capacity Qe of the energy store 3.

Qe = Co ^" (UcOmaχ-UcOmiπ) -

Here, Co is the acid capacity of the battery 3, U _c omax the open circuit voltage with fully charged battery and U _c omiπ the open circuit voltage at discharge.

In order to obtain a sufficiently accurate result for the capacity Qe of the battery 3, the parameters Co, U _c omiπ and U _c omax must be determined with sufficient accuracy. For the parameters Co and U _c omaχ this is easily possible in normal battery operation (near full charge). However, the parameter U _c omiπ can only be calculated accurately at low charge states <50%. Since these operating conditions occur only very rarely, the desired accuracy of the capacity calculation is rarely achieved. It is therefore proposed to use the parameter U _c omiπ on the basis of the parameters Rκo25 and vgr25, which are derived from the

Capacity of the battery 3 are dependent to calculate. The available capacity Qe of the battery 3 can be determined in the normal electrical system operation, without additional suggestions of the battery, the battery must not be discharged to low states of charge.

For the rest voltage of the battery 3 at discharge end U _c omiπ, for example, the following relationship can be applied: UcOmiπ_corr = U _c 0miπ + gl ^■ ΔR _K 25 - g2 ^■ Δvgr25.

ΔRκ25 and Δvgr25 are the changes of the parameters Rκ25 and vgr25 compared to a corresponding reference value, in particular the value in the new state of the battery 3. The factors gl and g2 are weighting factors.

The changes ΔRκ25 and Δvgr25 are here weighted in proportion to the accuracy with which the parameters R _K 25 and vgr25 were estimated by the state quantity and parameter estimator 1. The Kalman filter 1 outputs corresponding error variances P, which are included in the weighting factors gl and g2. The weighting factors gl and g2 can be expressed, for example, as follows:

g 1 ~ (P ₀ (RKO ₂₅ ) -P (RK ₀₂₅ )) / P ₀ (RK ₀₂₅ ) g2 ~ (P ₀ (vgr25) -P (vgr25)) / P ₀ (vgr25)

Where P ₀ (R _K025 ) and P ₀ (vgr25) are the initial error variances of the corresponding parameters, and P (R _K025 ) and P (vgr25) are the current error variances of the parameters R _K025 and vgr25 estimated by the _Kalman filter 1.

Instead of the deviation ΔRκ ₂₅ or Δvgr25, the deviation of one parameter of the double-layer capacitance C _DP between the positive electrode and the electrolyte could optionally or additionally also be used.

At low states of charge, in particular <50%, the parameter Ri _O 2 _{5 of} the internal resistance or its change in the calculation of the quiescent voltage U _c0 miπ at the end of discharge can also be included.

Claims

claims

1. A method for determining a battery size, in particular the capacity (Qe), of an energy store (3) with the aid of a state variable and parameter estimator (1) which uses operating variables (Ußatt, I Batt, Tßatt) of the energy store (3) the state variables ( Z) and parameters (P) of a mathematical energy storage model, characterized in that the sought battery size (Qe) of the energy store (3) is calculated as a function of at least one capacity-dependent parameter (Rκo25, vgr25).

2. The method according to claim 1, characterized in that the sought

Battery size (Qe) as a function of a minimum rest voltage (U _c omiπ) is calculated and the minimum rest voltage (U _c omiπ) as a function of the at least one capacity-dependent parameter (Rκo25, vgr25) is calculated

3. The method according to claim 1 or 2, characterized in that the sought battery size (Qe) as a function of a capacity-dependent parameter (Rκo2s) of the acid diffusion resistance (R _K ) of the energy store (3) is calculated.

4. The method according to any one of the preceding claims, characterized in that the sought battery size (Qe) as a function of a capacitance-dependent parameter (vgr25) of the passage resistance (R _DP ) between the electrolyte and an electrode of the energy store (3) is calculated.

5. The method according to any one of the preceding claims, characterized in that the sought battery size (Qe) as a function of

Deviation of the at least one capacity-dependent parameter (Rκo25, vgr25) is calculated from a reference value.

6. The method according to claim 5, characterized in that the deviation of the parameter (RkO25, vgr25) with a certain factor (g) is weighted.

7. The method according to any one of the preceding claims, characterized in that the sought battery size (Qe) is also calculated as a function of a capacity-dependent parameter (RiO2s) of the internal resistance (Ri).

8. The method according to any one of the preceding claims, characterized in that the sought battery size (Qe) of the energy store (3) as a function of a maximum rest voltage (U _c omax) calculated and the maximum rest voltage (U _c omax) by means of a learning algorithm (Kalman Filter) is adapted.

9. The method according to any one of the preceding claims, characterized in that the calculation is carried out by means of a Kalman filter.

10. The method according to any one of the preceding claims, characterized in that the battery size (Qe) is calculated as a function of a capacitance-dependent parameter of the double-layer capacitance (C _DP ) between the electrolyte and an electrode of the energy store (3).

11. Device for determining a battery size, in particular the capacity (Qe), of an energy store (3), comprising a state variable and parameter estimator (1), which from operating variables (Ußatt, I Batt, Tßatt) of the energy store (3) the state variables ( Z) and parameters (P) of a mathematical energy storage model, characterized in that a unit (2) is provided, by means of which the sought battery size (Qe) is calculated as a function of a capacity-dependent parameter (Rκo25, vgr25).