EP2076953A2

EP2076953A2 - Method for controlling the end of the discharge of a rechargeable battery

Info

Publication number: EP2076953A2
Application number: EP07858404A
Authority: EP
Inventors: Arnaud Delaille; Marion Perrin
Original assignee: Commissariat a lEnergie Atomique CEA
Current assignee: Commissariat a lEnergie Atomique et aux Energies Alternatives CEA
Priority date: 2006-10-13
Filing date: 2007-10-05
Publication date: 2009-07-08
Also published as: CN101523689A; JP5421111B2; US20090278502A1; CN101523689B; WO2008046980A3; FR2907272B1; JP2010506556A; WO2008046980A2; US8159190B2; FR2907272A1

Abstract

The invention relates to a method for controlling the end of the discharge of a rechargeable battery, that comprises determining (F1) the transition frequency (FHF) of the battery and determining (F2, F3) a discharge end criterion of said battery according to said transition frequency (FHF). The discharge end criterion is representative, e.g., of a predetermined maximal value (Pmax) of the slope of the transition frequency (FHF). The transition frequency (FHF) can be determined (F1) periodically when using the battery, the discharge being interrupted (F4) by regulation means according to said transition frequency (FHF).

Description

Method of managing the end of discharge of a rechargeable battery

Technical field of the invention

The invention relates to a method for managing the end of discharge of a rechargeable battery.

State of the art

Electrochemical accumulators, or rechargeable batteries, are used in many mobile applications, such as portable electronic, thermal, electric or hybrid vehicles, etc., or many stationary applications, such as telecommunication relay, electricity production from intermittent sources of energy, such as wind power or photovoltaics, etc., ensuring in each case an element that is as important as it is fragile.

In order to optimize their life span in these different applications, it is generally necessary to protect rechargeable batteries against any deep discharge, most of the time synonymous with premature degradation, in some cases unable to recharge. Also, in all applications in which the rechargeable batteries are likely to reach deep discharge states, Low Voltage Disconnect (LVD) thresholds are used to stop the discharge.

By way of example, as shown in the graph of FIG. 1, illustrating the voltage as a function of the quantity discharged from the battery, for three discharges made at three different current regimes, namely 0.5A (curve with the triangles), 5A (curve with the squares) and 1OA (curve with the circles), the voltage decreases in a conventional way to reach the value of 10, 8V, corresponding to the LVD low voltage threshold described above, the value at which the discharge must be stopped.

Once this voltage threshold is reached, the battery is disconnected. However, it should be noted that the voltage of the battery thus disconnected then increases immediately, due to the cancellation of overvoltages imposed by the current. If no precaution is taken, the discharge can be authorized again before being immediately interrupted, the resumption of the discharge signifying a new achievement of the low voltage threshold. To avoid this type of micro-discharges at low charge states, just as harmful for the battery as for the electronics ensuring the disconnection and reconnection of the battery, a new threshold of LVR reconnection voltage ("Low Voltage Reconnect" in English) is then used. However, the resumption of the discharge on this new criterion in voltage does not ensure that the battery has been recharged in the meantime to allow a new discharge, so that this process does in the end blur the harmful effect of micro-discharges without avoiding it altogether.

Moreover, an optimal adjustment of these voltage thresholds is very delicate. If these voltage thresholds take into account the battery technology or the sizing of the system, they are nevertheless maintained constant throughout the battery operation. Or ₁ the voltage of a battery depends on both of its technology, but also of its operating conditions, namely the charging current or discharging current, the relaxation period, temperature, and to finish his health , or state of wear. In the case of a battery which may include a series of elements, inhomogeneity between the elements may itself have an impact on the voltage measured across the complete battery. In other words, there is no direct relationship, or at least an approximate relationship, between the voltage of a battery and its state of charge. The voltage criteria used at present are therefore not able to fully appreciate the depth of discharge of rechargeable batteries by integrating all the contributions described above. Depending on the operating conditions or aging, these voltage thresholds therefore tend not to be ideal, and may even in some cases lead to premature degradation of the battery. In addition, the values of the cutoff thresholds for different discharge regimes are determined during the development of a new series of batteries by very empirical measurements, many, long and therefore expensive.

A conventional control structure of a battery 1 is shown diagrammatically in FIG. 2. In a conventional manner, a battery 1 supplies a load 2, via, for example, a power source 3. The charging cycles and discharging the battery 1 are managed by a regulator 4, connected to the terminals of the battery 1 and measuring, for example, the voltage across the battery 1. The controller 4 controls, according to S1, the charge of the battery 1 by via a first switch 5, in particular for disconnecting the battery after discharge, and control, according to S2, the energy source 3 via a second switch 6. To guard against such damage described above, the voltage thresholds used to stop the discharge are very largely overvalued within such a regulator described above, so as to ensure never to fall below the threshold of state of charge really critical e. For example, in the case of lead-acid batteries used in photovoltaic systems, the low voltage threshold of these batteries is generally close to 1.9V7 element (or 11.4V for a battery of 6 elements), whatever the conditions of wear or operation of the batteries. However, lead-acid batteries could most often be used up to a threshold close to 1.8V / element (or 10.8V for a battery of 6 elements), without any harmful consequences for their use. This results in a lack of storage energy, which can reach up to 10% of the battery capacity. Management of the end of discharge of such rechargeable batteries is not optimal.

Object of the invention

The object of the invention is to provide a method for managing the end of discharge of a rechargeable battery, which is efficient, simple to perform, from the exploitation of data specific to complex impedance measurements. , which allows a diagnosis of the state of the batteries, while taking into account the impact of the past and current conditions of use and the state of health of the batteries, and which makes it possible to avoid any inadvertent reconnections of the batteries fully discharged.

The method according to the invention is characterized in that it comprises the determination of the transition frequency of the battery and the determination of a criterion of end of discharge of the battery according to said transition frequency.

According to a particular embodiment of the invention, the transition frequency is determined periodically during use of the battery, the discharge being interrupted by regulating means as a function of said transition frequency. 007/001635

According to an alternative embodiment of the invention, said end of discharge criterion is determined from measurements of the transition frequency carried out during a calibration phase.

According to another embodiment of the invention, said end of discharge criterion is readjusted from measurements of the transition frequency made during a maintenance phase.

According to developments of the invention, said end of discharge criterion is representative of a sudden change of slope of the transition frequency during discharge, a predetermined maximum value of the transition frequency or a maximum value. predetermined slope of the transition frequency.

Brief description of the drawings

Other advantages and features will emerge more clearly from the following description of particular embodiments of the invention given by way of non-limiting example and represented in the accompanying drawings, in which:

FIG. 1 is a graph showing the discharged quantity of a battery as a function of the voltage, respectively at three different current regimes (0.5A, 5A and 10A), according to the prior art.

Figure 2 schematically shows a conventional control structure of a rechargeable battery according to the prior art.

FIG. 3 represents, in flowchart form, a particular embodiment of a method for managing the end of discharge of a rechargeable battery according to the invention. FIG. 4 illustrates the identification of the transition frequency F _HF in a Nyquist diagram, according to the prior art.

FIG. 5 is a graph showing the state of charge SOC of a battery as a function of the transition frequency F _HF , respectively at the three different current regimes according to FIG.

FIG. 6 is a graph illustrating the slope of the transition frequency as a function of the voltage, respectively at two different current regimes (0.5A and 5A), according to the invention.

Description of a preferred embodiment of the invention

With reference to FIG. 3, the method of managing the end of discharge of a rechargeable battery according to the invention can be achieved by a conventional control structure, as shown in FIG. 2, with a regulator making it possible to manage the load and discharging the battery, for example, by controlling switches.

In the particular embodiment shown in FIG. 3, the method of managing the end of discharge of a rechargeable battery consists in a first step F1 in determining the transition frequency F _HF of the battery.

By way of example, the determination of the transition frequency F _HF can be performed by dichotomy. Considering, for example, a lead-acid battery, with a frequency of transition F _HF in a frequency range of 3300HZ to 100Hz, and considering the imaginary part of the impedance measured at the frequency f, denoted lm (Z ( f)), and considering s the sign of this magnitude, we can then write: if s <0, then F _HF > f and if s ≥ 0, then F _HF ≤ f. The principle of the dichotomy then consists in making a first measurement, for example, at f1 = 1700 Hz, corresponding to the central point of the frequency range mentioned above. Following the sign of lm (Z (f1)) and the relation above, a new measurement will then be made, for example, at f2 = 2500 Hz, if s was less than 0, or f2 '= 900 Hz, if s was greater than 0. Following the sign of lm (Z (f2)), or respectively lm (Z (f2 ')), a new measurement will then be made, for example, at f3 = 2900 Hz, if s was less than 0, or f3 '= 2100 Hz, if s was greater than 0, or respectively f3 = 1300 Hz, if s was less than 0, or f3' = 500 Hz, if s was greater than 0, and so after.

In a general way, each new measure can finally be defined by the relation: fn + 1 = fn + x if s <0 and fn + 1 = fn - x if s> 0, with x From the 3300 Hz - 100 Hz range, seven measurements are then sufficient to obtain F _HF at 25 Hz, which is quite reasonable and sufficient.

Moreover, an estimation of the time of determination of the transition frequency F _HF by this method can be obtained by considering the realization of these seven successive measurements, all carried out at frequencies higher than 100 Hz. Considering that the measurements are moreover performed over five periods for each frequency studied, the realization of thirty-five measurements is then sufficient and much less than half a second, namely 0.35s if all the measurements are made at 100 Hz. This method by dichotomy therefore allows to monitor the frequency of transition F _HF every second, or over a longer time step.

In the field of impedance measurements, it was already known to determine the transition frequency F _HF . As shown in FIG. 4, illustrating the identification of the transition frequency F _HF in a diagram of FIG. Nyquist, the transition frequency F _HF conventionally corresponds to the frequency at which the complex impedance of the battery changes from an inductive behavior to a resistive behavior, that is to say the frequency with which the imaginary part of the complex impedance vanishes. . It is an impedance parameter measured in the high frequencies, continuously during the operation of the battery, whether at rest or in operation, in a completely non-intrusive manner.

However, all known current impedance measuring devices do not exploit this value of the F _HF transition frequency, but exploit the impedance values measured at various frequencies for battery health diagnostics. This is notably the case of document US 2003/204328, which describes an apparatus using a method for determining the state of health of a battery by measuring impedance values at various frequencies. Several recent publications, including the article "Monitoring state-of-charge of Ni-MH and Ni-Cd batteries using impedance spectroscopy" by A.Hammouche et al. (Journal of Power Sources, vol. 127, 2004, pp.105-11) and the article "Impedance measurements on lead-acid batteries for state-of-charge, state-of-health and cranking capabilities in hybrid and hybrid prognosis electric vehicles "by H. Blanke et al. (Journal of Power Sources, vol. 144, 2005, pp.418-425), as well as WO 2005/031380, also show the possibility of monitoring the state of charge of nickel-cadmium batteries, or lead batteries. -acid from this transition frequency F _HF . By way of example, as represented in FIG. 5, it is thus possible to determine the transition frequency F _HF as a function of the charge state SOC ("State On Charge" in English) of the battery, for the three same current regimes different from Figure 1, namely 0.5A, 5A and 10A.

However, no known document or apparatus refers to the possible use of the F _HF transition frequency as part of a FR2007 / 001635

regulation system, nor in the management of the end of discharge of the batteries from this criterion. The method according to the invention proposes, for its part, to use this transition frequency F _{HF1 in} order to manage the end of discharge of a rechargeable battery, the transition frequency F _HF being connected to the state of the active substance and being usable to stop the discharge depending on the actual state of the active ingredient.

In FIG. 3, after step F1 of determining the transition frequency F _HF , the management method includes determining, in a second step F2, the slope of the transition frequency F _HF . The slope is determined by the derivative of the transition frequency dF _HF / dt.

The derivative of the transition frequency dF _HF / dt can in particular be determined from the graph of FIG. 5. Indeed, this illustrates the transition frequency F _HF as a function of the state of charge SOC, when a constant current I discharge. However, the state of charge SOC is proportional to the current I multiplied by the time t. In FIG. 5, the current I being constant, the state of charge SOC is proportional to the time t. Thus, the derivative of each of the three curves shown in FIG. 5 thus corresponds, with a constant close which is the current I of discharge, to the slope of the transition frequency dF _HF / dt.

Then, in a third step F3, the management method consists of comparing the previously calculated slope with a maximum predetermined value Pmax, representative of the end of discharge criterion of the battery from which the battery must be disconnected. If the computed value of the slope is smaller than the maximum predetermined value Pmax, then the management method loops back before step F1 for determining the transition frequency F _HF . If the calculated value of the slope is greater than the predetermined maximum value Pmax, the management method then proceeds to the next step F4, namely the interruption of the discharge of the battery, by R2007 / 001635

through the regulator (Figure 2). The predetermined maximum value Pmax thus defines the end of discharge criterion, which is a function of the transition frequency, from which the battery must be disconnected.

By way of example, considering the graph shown in FIG. 6, which illustrates the value of the slope dF _HF / dt of the transition frequency F _HF as a function of the voltage U, at two different current regimes, namely 0.5A (curve with circles) and 5A (curve with crosses), it appears that the use of _HF slope dF / dt of the transition frequency F _HF can be quite conclusive in order to maintain the reliability of the end of discharge setpoint during aging of the battery.

Indeed, in the case of photovoltaic applications, it is common to stop the discharge on a fixed voltage shutdown setpoint, of the order of 11.5V, whatever the conditions of the discharge. For example, a stop of the discharge on a threshold of 11.5V to 5A (dashed lines in FIG. 6), which corresponds to a slope dF _HF / dt of the transition frequency F _HF of the order of 0.015, driven to maintain the same slope value dF _HF / dt of the transition frequency F _HF to stop the discharge at a threshold of the order of 11.25V to 0.5A.

In the embodiment described above, the transition frequency is determined (F1), for example, periodically during use of the battery in operation, the discharge then being interrupted (F4) by the regulator (FIG. function of the calculated transition frequency (F1).

In another variant embodiment not shown, the end of discharge criterion applied to the management method can be determined from measurements of the transition frequency F _HF made during a calibration phase of the battery. The end of discharge criterion corresponding to the calibration is then entered into the regulator (FIG. 2). The process consists then, as before, to determine the transition frequency F _HF (F1) and, for example, the slope (F2), the battery being disconnected (F4) when the end of discharge criterion calculated during the calibration is reached (F3) during battery operation.

In another variant embodiment not shown, the criterion of end of discharge can also be adjusted from the transition frequency F _HF measurements carried out during a maintenance phase of the battery. As before, the calculation of the transition frequency F _HF and the corresponding end of discharge criterion is then done during this maintenance phase and the battery is disconnected when the end of charge criterion is reached during the operation of the battery.

Furthermore, whether during a calibration phase or during a maintenance phase, the end of discharge criterion calculated as a function of the transition frequency F _HF and allowing the disconnection of the battery is preferably constituted by the voltage at the terminals of the battery measured by the regulator (FIG. 2), thus defining a fixed voltage stop instruction (FIG. 6).

In another non-illustrated embodiment of a method for managing the end of discharge of a rechargeable battery, the criterion of the end of discharge of the battery can be representative of a sudden change of slope of the transition frequency F _HF being discharged. The abrupt change of slope is defined by the second derivative of the transition frequency dF _HF ² / dt ² and the management method then consists in using this calculated value of the second derivative as a value representative of the end of discharge criterion, the battery being disconnected via the regulator (Figure 2) when this criterion is reached. In another variant embodiment, the criterion of end of discharge of the battery may be representative of a predetermined maximum value of the transition frequency F _HF . As before, the management method then consists in determining the transition frequency F _HF and comparing the value obtained with the predetermined maximum value. If this calculated value is lower than the predetermined maximum value, then the process loops back before the determination step F1 of the transition frequency F _HF and a new transition frequency F _HF is measured. If the calculated value is greater than the predetermined maximum value, then the end of discharge criterion has been reached and the controller interrupts the discharge by disconnecting the battery.

Whatever the embodiment described above, such a management method is very efficient and simple to perform, while allowing a diagnosis of the state of discharge of the batteries and taking into account the impact of the conditions of past and current use and health status of the batteries. Such a method also makes it possible to avoid any inadvertent reconnection of fully discharged batteries. The management of the discharge of a battery is optimal and the operation of the rechargeable battery, as well as its life, are optimized.

The invention is particularly applicable, thanks to the use of such end of discharge criterion according to the various embodiments of the management method described above, the determination and development of a new series of products. , the easy and inexpensive determination of end-of-discharge thresholds and its indication to users for all discharge regimes.

The invention also applies to the direct use in a regulator of such a new end of discharge criterion, for the adjustment of the end of discharge of the battery, so that the end of discharge takes consider the state of the battery in terms of aging or past use.

The invention is still applicable to the use, in a system comprising a battery and its regulator, of the new end of discharge criterion, so as to recalibrate the thresholds of said regulator during maintenance operations.

The invention is not limited to the various embodiments described above. The term battery used in the description above includes including both rechargeable batteries and rechargeable electrochemical accumulators. The management method described above applies in particular to any type of rechargeable batteries.

In general, all types of charge or discharge can be used, taking into account the type of battery and / or the type of application in which this battery is used.

Claims

4 R2007 / 001635References

1. A method for managing the end of discharge of a rechargeable battery, characterized in that it comprises determining (F1) the transition frequency (F _HF ) of the battery and determining an end criterion of discharge (F2, F3) of the battery according to said transition frequency (F _HF ).

2. Method according to claim 1, characterized in that the transition frequency (F _HF ) is determined periodically during use of the battery, the discharge being interrupted (F4) by regulating means according to said transition frequency (F _HF ).

3. Method according to claim 1, characterized in that said end of discharge criterion (F2, F3) is determined from measurements (F1) of the transition frequency (F _HF ) performed during a calibration phase.

4. Method according to claim 1, characterized in that said end of discharge criterion (F2, F3) is readjusted from measurements (F1) of the transition frequency (F _HF ) made during a maintenance phase.

5. Method according to one of claims 3 and 4, characterized in that said end of discharge criterion (F2, F3) is constituted by the voltage at the terminals of the battery.

6. Method according to any one of claims 2 to 5, characterized in that said end of discharge criterion (F2, F3) is representative of a sudden change in slope of the transition frequency (F _HF ) in the course of discharge.

7. Method according to any one of claims 2 to 5, characterized in that said end of discharge criterion (F2, F3) is representative of a predetermined maximum value of the transition frequency (F _HF ).

8. Method according to any one of claims 2 to 5, characterized in that said end of discharge criterion (F2, F3) is representative of a predetermined maximum value (Pmax) of the slope of the transition frequency (F _HF ).