• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of calibrating a battery for storing calibration data related to its state of health, characterized in that it comprises the following steps: E1: determination of at least one characteristic state of the battery, intermediate between its full charge and its full discharge, by detecting a peak in the evolution of the derivative of its unladen voltage during a charge or discharge of the battery, or at least one characteristic state at full or full charge discharging, or defined by the passage of a voltage threshold, respectively current, during a imposed current phase, imposed voltage respectively, during a charging or discharging of the battery; E2: definition of a reference state of the battery from this characteristic state; E4-E7: measurement of the no-load voltages OCVi of the battery in the reference state for several different SOHi health states and storing of these calibration data including at least pairs of real values state of health / voltage at no load (SOHi , OCVi) in an electronic memory.
    公开号:FR3020142A1
    申请号:FR1453408
    申请日:2014-04-16
    公开日:2015-10-23
    发明作者:Maxime Montaru
    申请人:Renault SAS；
    IPC主号:
    专利说明:

    [0001] The invention relates to a method for estimating the state of health of a battery. It also relates to a method of calibrating a battery, and a method of managing a battery. Finally, it also relates to a battery management system implementing this method for estimating the state of health of a battery. Battery management of the state of the art uses a representative indicator of the aging of a battery, often called by its English name of "State of Health" for health status, or more simply SOH. This indicator, sometimes also called state of life, is usually expressed as a percentage of the initial capacity at the new state of the battery, capacity measured in this initial state or provided by the manufacturer of the battery. SOH is commonly used in the diagnosis of a battery. Its evaluation is important for a good control of the functioning of the battery, as well as to better manage its end of life. According to a first approach, the real value of the SOH is measured by carrying out a capacitance test, consisting of a complete charge and then a complete subsequent discharge of the battery, under selected conditions of temperature and current. During this discharge, the quantity of charges restored is measured, which allows to deduce the actual capacity of the battery, and therefore its SOH. The advantage of this method is to obtain the real value of the SOH, since it relies on a measure of reality. However, its disadvantages are that it requires a significant duration, consumes energy, and requires intervention on the battery which is sometimes incompatible with its normal use, that is to say energy supply for the implementation. implementation of a certain concrete application: it then requires a pause of this application, for the implementation of the battery test. For this last reason, this first approach is considered intrusive since it generally requires stopping the normal use of the battery.
    [0002] To overcome the drawbacks of measuring real SOH, other less intrusive approaches exist, based on its estimation. As an example, a usual method for estimating the SOH is to follow the evolution of the resistance of the battery, or by extension to follow one or more impedance parameters of the battery. The wear of a battery is generally accompanied by an evolution of these parameters. The disadvantage of this method is not to directly measure the loss of capacity of the battery, but to estimate it from the evolution of a different parameter, while being very dependent on the measurement of the temperature. However, the evolution of capacity losses and increases in resistance, or impedances, do not follow generalized laws for all the batteries, or for all the aging conditions of the batteries. Thus, there are practical situations for which the increase in the resistance of a battery is sometimes negligible while the battery undergoes a significant loss of capacity, and vice versa.
    [0003] This estimation method therefore remains insufficiently reliable, and often complex because it requires additional calculations based for example on prior learning to try to overcome its failures.
    [0004] Thus, a general object of the invention is to provide a solution for estimating the SOH of a battery which does not include all or part of the disadvantages of the state of the art. More specifically, an object of the invention is to provide a battery SOH estimation solution, reliable, fast, non-intrusive.
    [0005] For this purpose, the invention is based on a method of calibrating a battery for storing calibration data related to its state of health, characterized in that it comprises the following steps: - determination of at least one state characteristic of the battery, intermediate between its full charge and its full discharge, by detecting a peak of the evolution of the derivative of its unladen voltage during a charge or discharge of the battery, or at least one characteristic state at its full charge or full discharge, or defined by the passage of a voltage threshold, respectively a current threshold, during a phase of imposed current, respectively imposed voltage, during a charging or discharging of the battery ; - definition of a reference state of the battery from this characteristic state; measuring the no-load voltages OCVi of the battery in the reference state for several different SOHi health states and storing these calibration data comprising at least pairs of real values health state / no-load voltage (SOHi, OCVi) in an electronic memory.
    [0006] The step of determining at least one characteristic state of the battery can take into account the derivative of its empty voltage with respect to the amount of charge accumulated by the battery or with respect to time.
    [0007] The reference state can be selected from a stabilized zone of the derivative of the no-load voltage near the peak of the detected characteristic state. The reference state can be defined by a shift of a fixed charge 01 of the charge accumulated by the battery with respect to its charge at the peak detected.
    [0008] The method of calibrating a battery may comprise the repetition of the following steps for several health states different from at least one battery: positioning of a battery in a state of health SOHi chosen and known; charging or discharging the battery, measuring and / or estimating the derivative of the no-load voltage of the battery and detecting the predefined peak of the derivative; continuing the charging or discharging of a predefined charge quantity from the predefined peak to reach the reference state of the battery, and measuring its empty voltage OCVi in this reference state; storage in an electronic memory comprising the calibration data of the pair of values comprising the value of the state of health SOHi and the empty voltage OCVi associated with the reference state of the battery. The method of calibrating a battery can implement slow charging or discharging phases of the battery to facilitate measurement of the no-load voltage. The no-load voltage of the battery can be obtained by one of the following steps: Measurement of the voltage at the terminals of the battery not connected to a charge after a relaxation time greater than one hour; or - Measurement of the voltage at the terminals of the battery not connected to a charge after a relaxation time less than or equal to one hour; or - Measurement of the voltage across the battery despite a very low current, emitted or received by the battery; or - Measuring the voltage across the battery under conditions different from the theoretical no-load voltage and implementing a correction of the measured voltage to estimate the no-load voltage; - Estimation of the no-load voltage of the battery from one or more measured electrical size (s) of the battery. The method of calibrating a battery may comprise a step of determining a law between the state of health of the battery and its empty voltage in the reference state, in particular by linear interpolation and / or the construction of an abacus associating a health status value of the battery with a value of no-load voltage of the battery at a reference state, based on the pairs of actual values state of health / no-load voltage (SOHi, OCVi).
    [0009] The method of calibrating a battery can be implemented from at least one battery of the family of batteries in the new state, and from the same battery in at least one state of aging.
    [0010] The invention also relates to a method for estimating the state of health of a battery, characterized in that it comprises a calibration phase implementing a calibration method as described above, and a phase of estimation of the state of health of the battery comprising the measurement of its unladen voltage in the reference state and the deduction of the estimate of its state of health from the calibration data obtained in the calibration and calibration phase. the measured no-load voltage. The second estimation phase may comprise all or some of the following preliminary steps, in order to determine its triggering: - Trigger command following a request from a user via a human machine interface of a device in which battery is used or a charging device of the battery; and / or - Automatic triggering according to a predefined frequency; and / or - Automatic triggering when the electrical conditions of the battery are favorable; - Automatic triggering during a charging phase of a battery; AND / OR can include the following preliminary steps: - A step of automatic detection of the electrical situation of the battery to detect the characteristic state; or - A step of modifying the configuration of the battery to position it near the reference state. The method of estimating the state of health of a battery can implement the following steps: measurement or estimation of the derivative of the voltage at no load during charging or discharging of the battery, detection of a peak of this derivative of the no-load voltage, - positioning of the battery at its reference state from the peak of the derivative of the no-load voltage, and measurement of the empty voltage of the battery, 25 - deduction of the state of health of the battery from the calibration data and the measured no-load voltage. The invention also relates to a computer medium readable by a management unit, characterized in that it comprises a registered computer program comprising computer program code means for implementing the state of the art estimation method. health of a battery as described above. The invention also relates to a device comprising at least one battery and a management unit, characterized in that the management unit implements a method for estimating the state of health of at least one battery such that previously described. The device can be a motor vehicle or a portable object such as a computer, a phone, a tablet or an electronic diary. The invention also relates to a device for calibrating a battery, characterized in that it implements a calibration method as described above. These objects, features and advantages of the present invention will be described in detail in the following description of a particular embodiment made in a non-limiting manner in relation to the attached figures among which: FIG. 1 represents several curves of evolution of the empty voltage OCV of a battery according to the charge transmitted to the battery during a charging phase for respectively several SOH health status values of the battery. FIG. 2 represents several evolution curves of the derivative of the function representing the empty voltage OCV of a battery by the charge transmitted to the battery during a charging phase as a function of this charge for respectively several values of SOH of the battery. drums. FIG. 3 represents the same curves as those of FIG. 2 but recalibrated with one another. FIG. 4 represents the same curves as those of FIG. 1 but recalibrated according to the registration defined by FIG. 3. FIG. 5 represents the relation between the state of health SOH of the battery and the empty voltage OCV of the battery. a reference state of the battery defined by the embodiment of the invention.
    [0011] FIG. 6 schematically represents a flowchart of a method for estimating the state of health of a battery according to one embodiment of the invention.
    [0012] The method according to the invention will exploit the empty voltage of a battery, also called OCV for the English name of "Open Circuit Voltage", which has the advantage of being independent of the temperature. This empty voltage is defined as the voltage measured at the battery terminals when it is not in use, and after a relaxation period following its last use, during which the voltage at its terminals evolves until it converges towards a stabilized value, which is the "no-load voltage". Note that this theoretical vacuum voltage is rarely used as defined above, because it would take a relatively long relaxation time, often several hours to reach a stabilized voltage value across the battery, comparable to the voltage This is why we are extending the definition of no-load voltage to values approaching it slightly, sometimes called "pseudo-vacuum voltage". for reasons of ease of exploitation of the associated solutions. Thus, the term of no-load voltage will be used later to also integrate the following values of voltage near a battery: - An empty voltage measured after a shortened relaxation time, for which the voltage at the terminals of the battery can not be considered to have converged to the theoretical value of no-load voltage; - A measured no-load voltage despite a very low current, emitted or received by the battery, for example in the context of a very slow charge phase, as at C / 25, where C would be the recommended rated load regime; - An estimated no-load voltage from one of the voltage measurements under conditions different from the vacuum voltage defined above, as in the two configurations mentioned above, by applying a correction to take account of this situation, in particular the voltage drop due to the impedance of the battery; - An empty voltage estimated from any other electrical measurement or electrical estimation of the battery; - An empty voltage, measured or estimated, by one of the approaches mentioned above, during a charge phase or discharge of a battery. One realizes that one does not obtain exactly the same value of voltage unladen at the terminals of a battery if this one was in phase of charge or discharge, with its relaxation. There are therefore potentially two values of no-load voltage, respectively charge and discharge. We will therefore consider one of the two values, or alternatively, an average value of the no-load voltages and discharge voltages. Thus, the method according to the embodiment of the invention, which will be described hereinafter, can be implemented with any one of the above approaches for the definition of the OCV vacuum voltage. However, it will be necessary to use the same definition of this empty voltage in all the stages of this method of calibration and estimation of the SOH to obtain a result as coherent and precise as possible.
    [0013] FIG. 1 comprises five curves 1 to 5 for the evolution of the empty voltage OCV (in V) of a battery as a function of the quantity of charge Qc (in Ah) accumulated by the battery during a charging phase, obtained for respective SOHi health states (i varying from 1 to 5) of the respective battery by 73%, 75%, 85%, 89% and 100%. These curves show that the value of the state of health SOH of a battery, for a given load, depends on its empty voltage. From the curves 1 to 5 of this FIG. 1, five curves 1 'to 5' are plotted in FIG. 2, respectively representing the values of the derivative of the vacuum voltage function as a function of the quantity of charge, for respectively the five curves 1 to 5. These curves all include a peak Pi (i ranging from 1 to 5). These different peaks Pi are slightly offset from each other, that is to say obtained for different levels of charge amounts Qc of the battery for different states of health SOH battery. These peaks locally represent maximums of these derivatives. Analyzes have shown that the peaks observed correspond to a particular physical state of the battery, in which a phase transition occurs at its two electrodes. This particular state (or characteristic state) moves with aging of the battery. It is easily detectable by simply observing curves 1 'to 5'. It will be used to define an intermediate reference state of the battery for the estimation of its state of health, as will be described below. The preceding conclusions will therefore be exploited for the implementation of an estimate of the state of health of a battery according to the following two principles: - This state of health will be deduced from the no-load voltage of the battery; - This state of health will be estimated for an intermediate charge state of the battery, between full charge and full discharge.
    [0014] A detailed embodiment of the invention will now be explained. This embodiment is based on a method for estimating the state of health of a battery, shown schematically in the diagram of FIG. 6, which comprises the following two phases: a first preliminary phase P1 is based on a method; calibration method for calculating calibration data, including health status reference values SOH of a certain battery as a function of its no-load voltage for a certain reference state; - A second estimation phase P2 of the SOH during the use of the battery, from the results of the calibration process. As a remark, in this first calibration phase, the SOH of a battery used for the calibration is calculated by a known and very precise method, in order to establish an accurate calibration with real values of SOH to achieve a great performance during the calibration. future estimates.
    [0015] According to the embodiment, the calibration method first comprises a first step E1 consisting in searching for at least one peak Pi of a curve of OCV / dQc as a function of the charge quantity Qc of a battery. For this, at least one of the curves 1 'to 5' is plotted, by implementing a load and a measurement of OCV values with a high frequency, that is to say a test procedure making it possible to obtain OCV values with sufficiently fine pitches compared to the variations of the ocV / dQ curve, to obtain the curve precisely and to ensure the correct detection of a peak. When the geometry of the selected peak is identified, a frequency of OCV measurements sufficient to detect the peak is chosen, but not necessarily as frequent to avoid any heavier process by too many OCV measurements. A second embodiment would consist in defining a detectable characteristic state by using the measurement of the voltage (respectively of the current) by the passage of a voltage threshold (respectively of the current) to an imposed current (respectively a voltage) for a certain temperature during a charge or discharge. The rest of the description will be detailed in the context of an embodiment associated with the detection of a peak.
    [0016] A second step E2 consists of determining a reference state of the battery from the identified peak (or peaks), which characterizes a particular state of the battery. This reference state is preferably not chosen at the level of the peak itself because, by definition, the value of the open-circuit voltage OCV has a greater variation with the quantity of charge Qc at this peak, which represents an unfavorable area. Indeed, the estimation method will be based on the measurement of the no-load voltage of the battery in the reference state and it is more favorable that this value is relatively stable around this reference state, to minimize errors. . For this reason, the reference state of the battery is chosen out of the peak, on an area for which the derivative has a low value, and therefore for which the OCV values are relatively stable. However, this reference state remains as close as possible to the peak identified, so as not to complicate its identification, which will be made from the recognition of the peak thereafter. Thus, the reference state is preferably distinct from a characteristic state detected by a peak, as explained above. To illustrate this choice of the reference state in the example illustrated, a figure 3 is constructed on which five curves 11 'to 15' are plotted, respectively corresponding to the curves 1 'to 5' of Figure 2, but shifted to obtain the superposition of their peaks Pi representing the particular intermediate state of the battery. Following the above explanations, it is chosen to determine the reference state of the calibration process outside the hatched area surrounding the peak, for which the value of the derivative is higher. Finally, in the embodiment described, the reference state is defined by a predetermined charging charge 01 of the battery from a peak Pi, which makes it possible to move away from the non-recommended zone, while remaining within proximity. The difference between a peak Pi and the reference state in the vicinity is preferably less than or equal to 5 Ah, or even 3 Ah. As an exemplary embodiment, it is also preferably greater than or equal to 2 Ah. FIG. 4 illustrates five curves 11 to 15 respectively corresponding to the curves 1 to 5 shifted according to the offset explained above, to allow the superposition of their reference state.
    [0017] The two preceding steps E1, E2 can be implemented for a single value of SOH of the battery, which is sufficient for the detection of a peak and the choice of a reference state. Alternatively, the relevance of a peak can be validated by the repetition of these steps for several different health states of the battery. Then, when these two preliminary steps have been performed, the method implements an iteration of the following steps (according to the number of distinct health states chosen and / or the number of batteries chosen), for several known health states of a battery: E4: positioning of the battery in a state of health SOHi chosen and known; E5: charge of the battery, measurement of the amount of charge and the no-load voltage, and especially of the derivative of the no-load voltage with respect to the load, and detection of the predefined peak; E6: continued charging of a value 01 from the identified peak to reach the reference state of the battery, and measurement of its voltage unladen; E7: storage in an electronic memory of the value of the SOH and the associated OCV vacuum voltage, these two values being obtained in the reference state of the battery. The pair (OCV, SOH) forms a reference data, a calibration data of the battery.
    [0018] As a remark, during the implementation of step E5, the initial state of the battery can be its fully discharged state, which makes it possible to measure the overall charge of the battery during its charging and to have an indication useful for confirming the recognition of the desired good peak, whose positioning with respect to the charge of the battery is in a relatively small known range. Alternatively, this charge could start at a state not completely discharged from the battery, but with a lower charge level than the peak (that is to say on the left of the peak sought in Figure 3).
    [0019] On the other hand, this charge is performed at a very slow rate, for example C / 25, which generates very low currents and allows a voltage measurement across the battery forming a pseudo vacuum voltage. Alternatively, a relaxation of the battery is implemented before each voltage measurement. In addition, this measurement of the no-load voltage and the measurement and / or estimation of the derivative of the no-load voltage is carried out periodically, according to a predefined period beforehand in a manner compatible with the detection of the desired peak. Finally, several pairs of real values (OCVi, SOHi) are thus obtained by the calibration method, for each state of health value SOH of the battery for which the previous steps E4 to E7 are implemented. By way of example, these pairs of real values are illustrated by points Ci on the graph of FIG.
    [0020] Then, the calibration method according to this embodiment comprises another step of elaboration of a law E8 between the SOH and the OCV of the battery in the reference state, to obtain a future estimate of the state of SOH health of the battery for any value of empty voltage OCV, from the few discrete values, represented by the points Ci, actually measured. In the example chosen, FIG. 5 shows, by the plot of line 7, a linear relationship between the SOH and the OCV at the reference state of the battery. The parameters making it possible to characterize this straight line, or more generally the law determined from the real points Ci, for example established by any mathematical interpolation method, can also be memorized with the calibration data. As a remark, this step E8 may alternatively be carried out later, during the use of the battery for example, not necessarily during the calibration phase.
    [0021] Other parameters, such as those for characterizing the charge of the battery implemented in the calibration phase, can also be stored as reference data, since the method will be implemented using preferably the same charging conditions. , for which the reference data are the most relevant. However, it is not mandatory to reproduce the same load conditions as those of the calibration, the calibration data can be used with any other load condition. In addition, the method chosen for calculating the no-load voltage of the battery is also memorized, in order to always reproduce the same calculation, as has been explained previously. In addition, the optimal frequency of the OCV measurements during charging can also be stored, for possible use in future estimation, and ensure the proper detection of the peak of the derivative of the no-load voltage. Finally, the approximate positioning of this peak can also be stored, for example within a range of charge values of the battery. Naturally, this calibration phase P1 can be carried out once (on a single battery) for a certain family of batteries, or several times, in particular on several batteries, to establish an average that is hoped for more accurate results. then applying thereafter to all the batteries of the family operating according to this same technology, when using them in a system.
    [0022] In addition, this calibration phase can be performed for a number of SOH values of the predetermined battery, which represents a compromise between the precision obtained, all the more high as the number of SOH is high, and the simplicity of the phase calibration, all the more simple and fast that the number of SOH is low. Subsequently, if we want a high accuracy, it is possible to increase the number of real SOH values considered. Advantageously, at least two points are calibrated (which may in particular be sufficient if the law is substantially linear), and preferably at least three, that is to say that the number of distinct SOH is greater than or equal to 3. Among the SOH considered, it is advantageous to integrate the new state of the battery and at least one aged state. In addition, to increase the accuracy of the method, it is also possible to store calibration data for several reference states of the same battery, for example distributed around the same peak and / or associated with different peaks. The method has been described by mentioning charging periods of the battery. Alternatively, the same steps could be implemented using discharge periods. The two approaches, charge and discharge, can be cumulated and / or combined during the calibration to offer more precision and flexibility thereafter. In addition, the method has used the variation of the empty voltage OCV of the battery according to the charge of the battery Qc but alternatively, it could rest in a similar manner on the variation of the open-circuit voltage OCV as a function of time and the derivative of the vacuum voltage with respect to time. In the latter case, it is then preferable to work at constant speed or by taking care of filtering the voltage signal (pass-pass filter). The process has also neglected the temperature. However, if a pseudo-OCV definition rather far from the theoretical no-load voltage is chosen, it may be useful to consider the temperature as a process variable, and implement the same calibration method for several different temperatures. The method is applicable to a single battery, that is to say an elementary battery in the form of an active storage part, for example based on chemical or capacitive components, arranged in a housing of which one first outer face forms the positive terminal or first current collector of the battery and a second outer face forms the negative terminal or second current collector. This housing has the function of maintaining and supporting the active part of the battery, and its sealing vis-à-vis the outside. It encloses a physical assembly that forms a fixed monolithic assembly, inseparable, whose electrical connection is unmodifiable for a user of the elementary battery, which sees an outgoing current equal to the current entering its two terminals.
    [0023] In a variant, the method is applicable to an electrical assembly of a plurality of elementary batteries, preferably having few dispersions and imbalances between them, and in particular to a battery pack, comprising an electrical assembly in series of several modules comprising elementary batteries in parallel. and / or in series. According to the embodiment, when the calibration phase has been established, once and for all, the method can use the resulting calibration data to estimate the SOH of any battery of the same type, throughout the lifetime of the the battery, during its use. If necessary, the calibration phase could be restarted, for example in case of doubt about the quality of the calibration data, but this is not necessary for the implementation of the SOH estimation method which will be described below. below.
    [0024] Thus, the method then uses a method for estimating the SOH of a battery by a second phase P2 for estimating the SOH during its normal use.
    [0025] This second phase P2 consists in deducing the state of health of a battery from the calibration data, as soon as the battery is in the reference state, from the measurement or estimation of its no-load voltage in this state. reference.
    [0026] This second phase P2 comprises the following preliminary steps, in order to trigger this second phase P2, according to the embodiment: An automatic detection step E12 of the electrical situation of the battery to detect that it is close to the reference state, for example from the knowledge of its state of charge or its no-load voltage; or alternatively - A modification step E13 of the battery configuration (charging or discharging) to position it near the reference state.
    [0027] Another prior triggering step El 1 of the second phase P2 may be based on one of the following steps: El 1 a: Trigger command following a request from a user via a human machine interface device in which the battery is used or a charging device of the battery; - E 1 1b: Automatic triggering according to a predefined frequency; - El 1 c: Automatic triggering when the electrical conditions of the battery are favorable, that is to say that it is close to the reference state; - Elld: automatic triggering during a charging phase of a battery (this charging phase being triggered independently, for example because the battery is heavily discharged). The steps El 1 and El 2 or El 3 can optionally be cumulated.
    [0028] Then, the method implements a step E14 which is then based on a charge period, or alternatively a discharge period, of the battery during which the no-load voltage is measured or estimated and / or the derivative of the measured no-load voltage. or estimated. This step is not necessarily performed under the same conditions as that of the calibration phase, in particular the charging conditions can here be nominal. Then, the method implements a step El detecting a peak of the derivative of the battery empty voltage, a positioning step E16 of the battery to its reference state, and measurement of the voltage at empty in the reference state, then a deduction step El 7 of its state of health from the calibration data and the measured no-load voltage. These steps are not detailed because they are similar to those explained for calibration.
    [0029] Naturally, the method may comprise a last step E19 of transmitting this estimate of SOH to an operator, via a human machine interface of an SOH estimator.
    [0030] The foregoing method has been described on the basis of the most common approach to define the state of health of a battery, which is to consider the loss of storage capacity of a quantity of charges. In a variant, the same method can be adapted for estimating the state of health of a battery according to another approach, for example by considering its loss in the possibility of restitution of energy with time. In such an approach, the adaptation of this process can be achieved by replacing the amounts of charges with amounts of energy.
    [0031] In addition, the invention can be implemented with different types of battery. It is particularly suitable for lithium ion batteries based on manganese or any type of technology exhibiting a similar behavior from the point of view of the invention.
    [0032] As mentioned previously, the curves illustrated were obtained during a charging phase of a battery. Conversely, we see the same phenomenon during a discharge phase. Thus, the embodiment could be inversely implemented by considering a discharge phase.
    [0033] One could also consider alternatively the derivative of the vacuum voltage by time, especially in the case of a charge or discharge at steady state.
    [0034] Finally, the embodiment of the invention has the following advantages: - It has the advantage of not requiring a full charge or discharge of the battery but to be able to proceed to estimate its state of health in a non-intrusive manner during the use of the battery, when it passes through its reference intermediate state, easy to identify; - The calculations implemented in the second phase P2 are simple, which allows their implementation without requiring a high computing power, and therefore compatible with an implementation within any device, such as a portable object possibly small; - The approach is easily generalizable to any type of battery, simply reproduce the calibration phase at least once for each type of battery; - The process shows in practice that it achieves a very high accuracy; - The method allows a rapid implementation, which makes it non-intrusive, because it can easily be implemented opportunistically, when the battery is close to its reference state, without disturbing the normal operation of the battery. The invention also relates to a device for estimating the state of health of a battery comprising a management unit implementing the method of estimating the state of health (SOH) of the battery as described above. , by hardware and / or software components. For this, the management unit comprises a computer, is associated with at least one electronic memory storing in particular the digital data from the first calibration phase, that is to say the calibration data, and a calculation software implementing all or part of the health status estimation (SOH) method. The invention also relates to such software as such. The device also comprises a man-machine interface for informing a user of the state of health of the battery, and interacting with him to define certain parameters of the method, for example the conditions for triggering the estimation. Finally, the device comprises at least one voltage and / or current and / or temperature measuring sensor connected by means of communication to the management unit. As a note, the battery management device, which implements the method of estimating its state of health, can be integrated with the battery itself. The invention also relates to a device equipped with a battery, which comprises such a device for estimating the state of health of the battery. The invention also relates to a battery calibration device, which makes it possible to implement the steps of detecting a peak of the derivative of the no-load voltage. This device therefore also comprises a computer and an electronic memory receiving the calibration data and other data to be memorized, for their future operation by an estimation device. This calibration device may also include a component implementing cycles of charges and discharges of a battery, to position it in a selected state of health. By way of nonlimiting example, the estimation method can thus be integrated within a motor vehicle, in particular an electric or hybrid vehicle, and it can be implemented during the normal operation of the motor vehicle. It can also be implemented in a battery charger or more specifically in a vehicle charging station. It can also be implemented within any portable object, such as a laptop, tablet, cell phone, e-notebook, etc. 5
    权利要求:
    Claims (16)
    [0001]
    REVENDICATIONS1. A method of calibrating a battery for storing calibration data related to its state of health, characterized in that it comprises the following steps: El: determining at least one characteristic state of the battery, intermediate between its full charge and its full discharge, by detecting a peak in the evolution of the derivative of its unladen voltage during a charge or discharge of the battery, or at least one characteristic state at full or full charge discharging, or defined by the passage of a voltage threshold, respectively current, during a imposed current phase, imposed voltage respectively, during a charging or discharging of the battery; E2: definition of a reference state of the battery from this characteristic state; E4-E7: measurement of the no-load voltages OCVi of the battery in the reference state for several different SOHi health states and storing of these calibration data including at least pairs of real values state of health / voltage at no load (SOHi , OCVi) in an electronic memory.
    [0002]
    2. A method of calibrating a battery according to the preceding claim, characterized in that the step of determining at least one characteristic state of the battery takes into account the derivative of its unladen voltage with respect to the amount of charge. accumulated by the battery or time.
    [0003]
    3. A method of calibrating a battery according to one of the preceding claims, characterized in that the reference state is chosen in a stabilized zone of the derivative of the vacuum voltage near the peak of the detected characteristic state.
    [0004]
    4. A method of calibrating a battery according to the preceding claim, characterized in that the reference state is defined by a shift of a fixed charge 01 of the charge accumulated by the battery relative to its load at the peak detected.
    [0005]
    5. A method of calibrating a battery according to one of the preceding claims, characterized in that it comprises the repetition of the following steps for several health states separate from at least one battery: E4: positioning a battery in a state of health SOHi chosen and known; E5: charge or discharge of the battery, measurement and / or estimation of the derivative of the no-load voltage of the battery and detection of the predefined peak of the derivative; E6: charging or discharging a predefined charge quantity from the predefined peak to reach the reference state of the battery, and measuring its empty voltage OCVi in this reference state; E7: storage in an electronic memory comprising the calibration data of the pair of values comprising the value of the state of health SOHi and the empty voltage OCVi associated with the reference state of the battery.
    [0006]
    6. A method of calibrating a battery according to one of the preceding claims, characterized in that it implements slow charge or discharge phases of the battery to facilitate the measurement of the no-load voltage.
    [0007]
    7. A method of calibrating a battery according to one of the preceding claims, characterized in that the no-load voltage of the battery is obtained by one of the following steps: - Measurement of the voltage at the terminals of the battery not connected at a charge after a relaxation time greater than one hour; or - Measurement of the voltage at the terminals of the battery not connected to a charge after a relaxation time less than or equal to one hour; or - Measurement of the voltage across the battery despite a very low current, emitted or received by the battery; or - Measuring the voltage across the battery under conditions different from the theoretical no-load voltage and implementing a correction of the measured voltage to estimate the no-load voltage; - Estimation of the no-load voltage of the battery from one or more measured electrical size (s) of the battery.
    [0008]
    8. A method of calibrating a battery according to one of the preceding claims, characterized in that it comprises a step (E8) for determining a law between the state of health of the battery and its voltage unladen in the reference state, in particular by linear interpolation and / or the construction of an abacus associating a health status value of the battery with a value of no-load voltage of the battery at a reference state, from the pairs actual health status / no-load voltage values (SOHi, OCVi).
    [0009]
    9. A method of calibrating a battery according to one of the preceding claims, characterized in that it is implemented from at least one battery of the family of batteries in the new state, and from the same battery in at least one state of aging.
    [0010]
    10. A method for estimating the state of health of a battery, characterized in that it comprises a calibration phase (P1) implementing a calibration method according to one of the preceding claims, and a phase of estimation (P2) of the state of health of the battery comprising the measurement of its unladen voltage in the reference state and the deduction of the estimate of its state of health from the calibration data obtained in the phase calibration (P1) and the measured no-load voltage.
    [0011]
    11. Method for estimating the state of health of a battery according to the preceding claim, characterized in that the second estimation phase (P2) comprises all or part of the following preliminary steps, in order to determine its triggering: - (E11 a) Trigger command following a request from a user via a human machine interface of a device in which the battery is used or a charging device of the battery; and / or - (E11 b) Automatic triggering according to a predefined frequency; and / or - (E11c) Automatic triggering when the electrical conditions of the battery are favorable; - (E11d): automatic triggering during a charging phase of a battery; AND / OR in that it comprises the following preliminary steps: an automatic detection step (E12) of the electrical situation of the battery to detect the characteristic state; or - A step of modifying (E13) the configuration of the battery to position it near the reference state.
    [0012]
    12. A method for estimating the state of health of a battery according to claim 10 or 11, characterized in that it implements the following steps: E14 - measurement or estimation of the derivative of the vacuum voltage during a charge or discharge of the battery, E15 - detection of a peak of this derivative of the no-load voltage, E16 - positioning of the battery at its reference state from the peak of the derivative of the no-load voltage, and measuring the no-load voltage of the battery, E17 - deduction of the health status of the battery from the calibration data and the measured no-load voltage.
    [0013]
    13. Computer support readable by a management unit, characterized in that it comprises a registered computer program comprising computer program code means for implementing the method for estimating the state of health of a battery according to one of claims 10 to 12.
    [0014]
    14. Device comprising at least one battery and a management unit, characterized in that the management unit implements a method for estimating the state of health of at least one battery according to one of the claims. 10 to 12.
    [0015]
    15. Device according to the preceding claim, characterized in that it is a motor vehicle or a portable object such as a computer, a phone, a tablet or an electronic agenda.
    [0016]
    16. Device for calibrating a battery, characterized in that it implements a calibration method according to one of claims 1 to 9. 30
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3132273B1|2018-06-06|Method for estimating the state of health of a battery
    EP2397863B1|2012-12-12|System for monitoring the status of a battery
    EP3080625B1|2020-03-11|Method for estimating the state of health of a battery
    EP2410346B1|2017-07-12|Method of determining a parameter of at least one accumulator of a battery
    EP3014292B1|2017-05-03|Method for evaluating the state of charge of a battery
    WO2015015083A1|2015-02-05|Estimation of the state of deterioration of an electric battery
    EP3052952B1|2017-11-15|Method for monitoring a li-ion battery and monitoring device for implementation of same
    FR2976743A1|2012-12-21|METHOD FOR MANAGING AND DIAGNOSING A BATTERY
    WO2017050944A1|2017-03-30|Method and device for determining a state-of-health indicator for a lithium battery
    FR2977678A1|2013-01-11|Method for diagnosing battery, involves charging battery to reach predetermined maximum voltage for specific time in phase, and calculating state of health of battery from measured time and charging time
    WO2012114036A1|2012-08-30|On-board device for estimating the ageing of a power battery of a motor vehicle, and corresponding method
    WO2017108536A1|2017-06-29|Method for assessing the state of charge of a battery
    FR3090117A1|2020-06-19|ESH ESTIMATION AND SOC ESTIMATE OF AN ELECTROCHEMICAL ELEMENT
    EP2697859B1|2015-03-04|Method for optimally charging an electrochemical battery
    EP3667345B1|2021-04-21|Method for determining the state of health of the cells of a battery
    EP3221710B1|2018-10-10|Method for characterising an accumulator
    EP3234622B1|2018-12-12|Method and system for charging and balancing a battery module and/or pack comprising electrochemical elements
    EP3080624B1|2020-03-11|Method of estimating the state of health of a battery
    EP3948316A1|2022-02-09|Method for initialising the charge state of a battery
    WO2021064030A1|2021-04-08|Method for determining the state of health of a lithium-ion battery
    FR2971855A1|2012-08-24|Device for monitoring of charging state of lithium-ion battery embarked on e.g. hybrid propulsion motor vehicle, has calculation unit configured periodically to reset frequency table during service life of battery
    FR3045218A1|2017-06-16|DETERMINATION OF PARAMETERS OF A DYNAMIC MODEL FOR AN ELECTROCHEMICAL BATTERY CELL
    同族专利:
    公开号 | 公开日
    JP6734784B2|2020-08-05|
    EP3132273B1|2018-06-06|
    US10989761B2|2021-04-27|
    EP3132273A1|2017-02-22|
    WO2015158813A1|2015-10-22|
    US20170038436A1|2017-02-09|
    CN106461732B|2019-12-20|
    FR3020142B1|2016-05-13|
    CN106461732A|2017-02-22|
    KR20160144437A|2016-12-16|
    JP2017514127A|2017-06-01|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20020030494A1|2000-07-28|2002-03-14|Honda Giken Kogyo Kabushiki Kaisha|Remaining charge detection device for power storage unit|
    US20120035873A1|2008-08-08|2012-02-09|Kang Jung-Soo|Apparatus and method for estimating state of health of battery based on battery voltage variation pattern|
    FR2955670A1|2010-01-27|2011-07-29|Commissariat Energie Atomique|METHOD FOR DIAGNOSING THE HEALTH CONDITION OF A BATTERY|
    US20130166234A1|2011-12-26|2013-06-27|Industrial Technology Research Institute|Method for estimating battery degradation|FR3090114A1|2018-12-17|2020-06-19|Electricite De France|Battery health|US5780994A|1997-03-21|1998-07-14|Securaplane Technologies, L.L.C.|Detection of inflection point in secondary-battery charging process by matching voltage response to first derivative of battery's characteristic curve|
    WO1998054588A1|1997-05-27|1998-12-03|Koninklijke Philips Electronics N.V.|Battery-powered electrical device|
    US6630814B2|2000-12-19|2003-10-07|Telefonaktiebolaget Lm Ericsson |Method and apparatus for calibrating a rechargeable battery|
    FR2942545B1|2009-02-24|2012-08-03|Helion|METHOD FOR DETERMINING A HEALTH STATUS OF AN ELECTROCHEMICAL DEVICE|
    JP5419831B2|2010-09-06|2014-02-19|カルソニックカンセイ株式会社|Battery degradation degree estimation device|
    US8531158B2|2010-11-01|2013-09-10|GM Global Technology Operations LLC|Method and apparatus for assessing battery state of health|
    CN102445645A|2011-09-22|2012-05-09|西北工业大学|Device and test method for measuring solar cell characteristic|
    JP2013247003A|2012-05-28|2013-12-09|Sony Corp|Charge control device for secondary battery, charge control method for secondary battery, charged state estimation device for secondary battery, charged state estimation method for secondary battery, deterioration degree estimation device for secondary battery, deterioration degree estimation method for secondary battery, and secondary battery device|
    EP3011351B1|2013-06-19|2017-08-16|Volvo Truck Corporation|Method for estimating state of health of a battery in a hybrid vehicle|
    FR3010532B1|2013-09-11|2017-06-09|Commissariat Energie Atomique|METHOD, DEVICE AND SYSTEM FOR ESTIMATING THE CHARGE STATE OF A BATTERY|KR102356475B1|2015-02-06|2022-01-27|삼성전자주식회사|Method and apparatus for estimating state of battery based on battery charging voltage data|
    DE102017102668A1|2017-02-10|2018-08-16|Dr. Ing. H.C. F. Porsche Aktiengesellschaft|Method and device for monitoring a temperature of a battery system|
    US11215675B2|2017-04-17|2022-01-04|The Regents Of The University Of Michigan|Method to estimate battery health for mobile devices based on relaxing voltages|
    KR101792975B1|2017-04-25|2017-11-02|한국기술교육대학교 산학협력단|Method for Predicting State of Health of Battery Based on Numerical Simulation Data|
    TWI621866B|2017-05-05|2018-04-21|致茂電子股份有限公司|Test method for semi-finished product of battery core|
    CN107831444B|2017-10-26|2019-12-03|哈尔滨工业大学|A kind of health state of lithium ion battery estimation method|
    KR101930646B1|2017-11-22|2019-03-11|주식회사 포스코아이씨티|Apparatus and Method for Estimating Capacity of Battery Using Second Order Differential Voltage Curve|
    KR102244140B1|2017-12-21|2021-04-22|주식회사 엘지화학|Method and apparatus for calibrating state of charge of a battery|
    KR20190075684A|2017-12-21|2019-07-01|삼성전자주식회사|An battery monitoring apparatus detecting charge balance of a battery and a method thereof|
    CN110549909B|2018-03-30|2021-06-18|比亚迪股份有限公司|SOH calculation method and device of power battery pack and electric vehicle|
    CN108535662A|2018-05-22|2018-09-14|华霆（合肥）动力技术有限公司|Cell health state detection method and battery management system|
    WO2020046019A1|2018-08-29|2020-03-05|주식회사 엘지화학|Battery management device, battery managing method, battery pack and electric vehicle|
    KR102259415B1|2018-08-29|2021-06-01|주식회사 엘지에너지솔루션|Battery management apparatus, battery management method, battery pack and electric vehicle|
    CN110988690B|2019-04-25|2021-03-09|宁德时代新能源科技股份有限公司|Battery state of health correction method, device, management system and storage medium|
    CN110333450B|2019-04-30|2021-11-19|蜂巢能源科技有限公司|Battery open-circuit voltage estimation method and system|
    DE102019111976A1|2019-05-08|2020-11-12|TWAICE Technologies GmbH|Determining the capacity of batteries|
    CN110967631B|2019-05-17|2021-03-30|宁德时代新能源科技股份有限公司|SOH correction method and apparatus, battery management system, and storage medium|
    DE102019126245A1|2019-09-30|2021-04-01|Dr. Ing. H.C. F. Porsche Aktiengesellschaft|System and method for determining the functional status and / or health status of an electric battery|
    US20210223326A1|2020-01-20|2021-07-22|Southwest Research Institute|System and method for estimating battery state of health|
    WO2022034671A1|2020-08-13|2022-02-17|ＴｅｒａＷａｔｔ Ｔｅｃｈｎｏｌｏｇｙ株式会社|Deterioration state estimation device, deterioration state estimation method, program, and power supply device for anode-free lithium battery equipped with same|
    法律状态:
    2015-04-21| PLFP| Fee payment|Year of fee payment: 2 |
    2015-04-30| PLFP| Fee payment|Year of fee payment: 3 |
    2016-04-21| PLFP| Fee payment|Year of fee payment: 3 |
    2017-04-19| PLFP| Fee payment|Year of fee payment: 4 |
    2018-04-20| PLFP| Fee payment|Year of fee payment: 5 |
    2019-04-18| PLFP| Fee payment|Year of fee payment: 6 |
    2021-01-15| ST| Notification of lapse|Effective date: 20201209 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1453408A|FR3020142B1|2014-04-16|2014-04-16|METHOD FOR ESTIMATING THE HEALTH CONDITION OF A BATTERY|FR1453408A| FR3020142B1|2014-04-16|2014-04-16|METHOD FOR ESTIMATING THE HEALTH CONDITION OF A BATTERY|
    KR1020167031226A| KR20160144437A|2014-04-16|2015-04-16|Method for estimating the state of health of a battery|
    PCT/EP2015/058252| WO2015158813A1|2014-04-16|2015-04-16|Method for estimating the state of health of a battery|
    US15/303,963| US10989761B2|2014-04-16|2015-04-16|Method for estimating the state of health of a battery|
    CN201580020056.6A| CN106461732B|2014-04-16|2015-04-16|Method for estimating state of health of battery|
    JP2016562914A| JP6734784B2|2014-04-16|2015-04-16|How to estimate battery health|
    EP15718186.8A| EP3132273B1|2014-04-16|2015-04-16|Method for estimating the state of health of a battery|
    [返回顶部]