Method and device for detecting a failure of a lithium-ion battery cell

Abstract

The invention relates to a method (100) for detecting a failure of a cell of a Lithium-Ion battery. The method comprises, for at least one CV phase of a CC-CV charging cycle of the cell: a collection (110) of several measurements forming a floating current signal, a derivation (120) of the floating current signal, a decomposition into empirical modes (130) of the derivative of the floating current, a determination (150) of an incidence value representative of the state of health of the cell, for the CV phase considered, from the intrinsic components obtained by the decomposition, an evaluation (160) of a failure criterion of the cell as a function of the incidence value. Figure for the abstract: Fig. 1

Description

Title of the invention: Method and device for detecting a failure of a Lithium-Ion battery cell Field of invention

[0001] The present invention belongs to the field of management of “Lithium-Ion” type batteries. More particularly, a method and a device are proposed for detecting a failure of a Lithium-Ion battery cell, in particular with the aim of preventing an accident (explosion, fire or other) during the life of the battery. State of the art

[0002] There are currently various methods for detecting failures of lithium batteries.

[0003] Some methods are based on measurements of electromechanical properties of the battery (vibration, acoustic waves, etc.). These measurements can be used to identify early signs of failure, such as cracks, delaminations and manufacturing defects.

[0004] Some methods rely on measurements of thermal properties of the battery, such as temperature, thermal conductivity and thermal capacity. These measurements can be used to identify abnormal heating of the battery.

[0005] A disadvantage of these methods is that the failure is detected quite late. These methods therefore do not always make it possible to satisfactorily prevent an accident linked to a battery failure.

[0006] Other methods are based on measurements of the electrochemical properties of the battery, such as voltage, current, resistance or capacity.

[0007] Among these electrochemical methods, a large number are based on voltage signals measured at the battery level. Examples include patent applications US 2017 / 146608 A1 and US 2022 / 0381849 A1. Patent application EP 3324197 A1 describes a method for determining the state of health of a battery cell based on a ratio between a charge variation and a current difference measured between two instants of a CV phase (constant voltage charging phase) of a CC-CV cycle (charging cycle comprising a constant current charging phase followed by a constant voltage charging phase). The reliability of these methods for detecting a failure is, however, not always fully satisfactory. In particular, these methods generally do not make it possible to anticipate a failure of a battery cell sufficiently early. Statement of the invention

[0008] The present invention aims to remedy all or part of the drawbacks of the prior art, in particular those set out above.

[0009] For this purpose, and according to a first aspect, the present invention proposes a method for detecting a failure of a cell of a Lithium-Ion battery. The method comprises, for at least one “constant voltage” phase (CV phase) of a “constant current - constant voltage” charging cycle (CC-CV charging cycle) of the cell: - a collection of a plurality of current measurements carried out at the cell level during said at least one CV phase, said plurality of measurements forming a “floating current” signal, - a derivation of the floating current signal to obtain a floating current derivative signal, - a decomposition into empirical modes of the floating current derivative signal in order to obtain a representation in the form of a sum of a residual signal and one or more intrinsic components, - a determination of an incidence value representative of a state of health of the cell, for said at least one CV phase, from the intrinsic components thus obtained, - an evaluation of a criterion for detecting a failure of the cell as a function of the incidence value of said at least one CV phase.

[0010] In particular embodiments, the invention may further comprise one or more of the following characteristics, taken in isolation or in all technically possible combinations.

[0011] In particular modes of implementation, the determination of the incidence value, for said at least one CV phase, comprises: - a calculation of an energy for each intrinsic component, - a calculation of a total intrinsic energy equal to a sum of the energies of the intrinsic components, - a determination of the incidence value as a function of the total intrinsic energy thus calculated.

[0012] In particular embodiments, the determination of the incidence value, for said at least one CV phase, comprises: - a calculation of a spectral density for each intrinsic component, - a calculation of a total intrinsic spectral density equal to a sum of the spectral densities of the different intrinsic components, - a determination of the incidence value as a function of the total intrinsic spectral density thus calculated.

[0013] In particular embodiments, for each intrinsic component, the spectral density of the intrinsic component is calculated from a Hilbert transform of the intrinsic component.

[0014] The proposed method is clearly distinguished from conventional methods in that it is based on the analysis of the derivative of the floating current during a CV phase of a CC-CV charging cycle. Nothing suggests at first glance that this signal contains relevant information for monitoring the health status of the cell.

[0015] The decomposition into empirical modes is particularly well suited to the analysis of this signal. This decomposition also has the advantage of being relatively fast and not very demanding in terms of computing capacity.

[0016] Monitoring the total intrinsic energy or the total intrinsic spectral density of the cell allows for early and efficient failure detection.

[0017] In particular embodiments, the method further comprises, for said at least one CV phase, an estimation of a statistical reliability of the CV phase, as a function of the intrinsic components of the floating current derivative signal. The CV phase is filtered if it is deemed unreliable.

[0018] In particular embodiments, the statistical reliability of the CV phase is estimated as a function of an entropy calculated for a sum of the intrinsic components of the floating current derivative signal.

[0019] In particular embodiments, the statistical reliability of the CV phase is estimated by comparing the total intrinsic energy with a predetermined energy threshold, or with the total intrinsic energies calculated for all or part of previous CV phases.

[0020] This step of filtering CV phases deemed statistically unreliable makes it possible to exclude CV phases presenting aberrant incidence values.

[0021] In particular embodiments, the evaluation of the criterion for detecting a failure of the cell comprises a comparison of the incidence value with a predetermined incidence threshold, for said at least one CV phase.

[0022] In particular embodiments, the evaluation of the criterion for detecting a failure of the cell comprises a comparison of a slope of a linear interpolation carried out for a set of several incidence values ​​corresponding to several consecutive CV phases with a predetermined slope threshold.

[0023] In particular embodiments, the evaluation of the criterion for detecting a failure of the cell comprises a comparison of the incidence value for said at least one CV phase with an incidence value of a previous CV phase.

[0024] These different criteria can be used, individually or in combination, to detect a cell failure.

[0025] In particular modes of implementation, when a failure is detected, the method further comprises a verification whether the detected failure is linked to the environment in which the cell has evolved.

[0026] In particular embodiments, the verification whether the detected failure is linked to the environment comprises a comparison, for a given period comprising said at least one CV phase, of incidence values ​​determined for the cell during said period with incidence values ​​determined for at least one other cell subjected to the same environment during said period.

[0027] In particular embodiments, the verification whether the detected failure is linked to the environment comprises a comparison, for a given period comprising said at least one CV phase, of environmental measurements carried out during said period with a predetermined threshold.

[0028] It may indeed be advantageous to know whether a detected failure is linked to the environment in which the cell has evolved. If the failure is linked to the environment, it is possible to take corrective action at the level of the cell's environment, and the failure will potentially have a limited impact over time. A failure that is not linked to the environment is potentially more serious since it predicts an intrinsic failure of the cell, such as for example a failure linked to a manufacturing defect or early degradation of the cell. An intrinsic failure of the cell may require the cell to be made safe to avoid a risk of accident (swelling of the cell, risk of rupture of a casing enclosing the cell, thermal runaway, risk of fire or explosion, etc.).

[0029] According to a second aspect, the present invention provides a device for detecting a failure of a cell of a Lithium-Ion battery. The device comprises: - a battery management system configured to provide current measurements made at the cell level during a “constant voltage” phase (CV phase) of a “constant current - constant voltage” charge cycle (CC-CV charge cycle) of the cell, - a computing unit connected to the battery management system, said computing unit being configured to implement a method according to any one of the implementation modes previously described. Presentation of figures

[0030] The invention will be better understood on reading the following description, given by way of non-limiting example, and made with reference to Figures 1 to 15 which represent:

[0031] [Fig-1] a schematic representation of the main stages of an example of implementation implementing the method according to the invention for detecting a failure of a cell of a Lithium-Ion battery,

[0032] [Fig.2] a schematic representation of a particular mode of implementation of the step of determining an incidence value,

[0033] [Fig.3] a schematic representation of another particular mode of implementing the step of determining an incidence value,

[0034] [Fig.4] a graph representing the current flowing in a Lithium-Ion battery cell during four successive CC-CV charge cycles,

[0035] [Fig.5] a graph representing the floating current flowing in a Lithium-Ion battery cell during a CV phase of a CC-CV charge cycle of the cell,

[0036] [Fig.6] a graph representing the derivative of the floating current shown in [Fig.5],

[0037] [Fig.7] a graph representing a decomposition into empirical modes of the floating current derivative represented in [Fig.6],

[0038] [Fig.8] a schematic representation of a particular mode of implementation of the step of evaluating a criterion for detecting a failure of the cell,

[0039] [Fig.9] a first example of incidence values ​​determined successively for seven CV phases,

[0040] [Fig. 10] a second example of incidence values ​​determined successively for seven CV phases,

[0041] [Fig. 11] a third example of incidence values ​​determined successively for seven CV phases,

[0042] [Fig. 12] a fourth example of incidence values ​​determined successively for seven CV phases,

[0043] [Fig. 13] a graph representing the evolution over time of the incidence value of a battery cell, as well as the evolution over time, during the same period, of the temperature experienced by the cell,

[0044] [Fig. 14] a graph representing another example of the evolution over time of the incidence value of a battery cell, as well as the evolution over time, during the same period, of the temperature experienced by the cell,

[0045] [Fig. 15] a schematic representation of a device according to the invention for detecting a failure of a cell of a Lithium-Ion battery.

[0046] In these figures, identical references from one figure to another designate identical or similar elements. For reasons of clarity, the elements represented are not necessarily on the same scale, unless otherwise stated. Detailed description of the invention

[0047] As previously indicated, the present application provides a method and a device for detecting a failure of a cell of a Lithium-Ion battery.

[0048] A charging cycle of a Lithium-Ion battery cell typically comprises two phases: a first constant current charging phase, or CC phase (for "Constant Current" in English) and a second constant voltage charging phase, or CV phase (for "Constant Voltage" in English). This is then referred to as a CC-CV charging cycle. In the present invention, the current flowing in the cell during the CV phase is of interest. This current is generally referred to as the "floating current".

[0049] Although conventionally the first charging phase (CC phase) is carried out at constant current, more elaborate charging schemes may exist during this first phase, with a change in the current for example to maximize the charging speed while remaining within the charging range compatible with the accumulator. The end of the first charging phase is then marked by the reaching of a voltage threshold which triggers the switch to the second charging phase at constant voltage (CV phase). For simplicity, these particular cases of the first charging phase will be considered as also covered by the term CC-CV charging “constant current - constant voltage” used in this application.

[0050] [Fig.4] is a graph representing four successive CC-CV charge cycles of a Lithium-Ion battery cell. The current flowing through the cell is represented on the ordinate (in amperes) and the time is represented on the abscissa (in seconds). Each cycle has a CC 31 phase and a CV 32 phase. As illustrated in the graph in [Fig.4], the current flowing through the cell during the CC 31 phase is approximately constant, and it exhibits an exponential decay during the CV 32 phase. [Fig.5] is a graph showing in more detail the CV 32 phase of a CC-CV charge cycle of the cell.

[0051] [Fig.l] schematically represents the main steps of an example of implementation of a method 100 according to the invention for detecting a failure of a cell of a Lithium-Ion battery.

[0052] As illustrated in [Fig.l], the method 100 comprises the following steps, for at least one CV phase of a CC-CV charge cycle of the cell: - a collection 110 of several floating current measurements during the CV phase, - a 120 derivation of the floating current thus obtained, - a decomposition into empirical modes 130 (EMD for “Empirical Mode Decomposition » in English) of the derivative of the floating current, - a determination 150 of an incidence value representative of a state of health of the cell from the intrinsic components obtained by the EMD decomposition, - an evaluation 160 of a criterion for detecting a cell failure in function of the incidence value thus determined.

[0053] These steps can be implemented for the CV phase of one or more CC-CV charging cycles undergone by the cell during its life. They can, for example, be implemented at each new CV phase of the cell, in order to systematically monitor the state of health of the cell. However, nothing would prevent, in a variant, implementing these steps only for a subset of all the charging cycles of the cell, for example each time a certain number of charging cycles has been carried out, or with a predetermined time frequency. These steps can also be implemented sporadically during the life of the cell.

[0054] The floating current measurements during the CV phase are for example carried out by a battery management system (BMS) connected to the cell. The measurements are for example carried out with an acquisition frequency of between one and sixty seconds, for an acquisition duration of ten to sixty minutes. However, nothing would prevent the measurements from being carried out with a different acquisition frequency, and / or for a different acquisition duration. It is advantageous to use a number of measurements of between thirty and fifty per CV phase (using a larger number of measurements does not necessarily imply a significant improvement in the method; using a smaller number of measurements may, however, limit the performance of the method). All of the measurements collected during the collection step 110 form a “floating current” signal. The graph in [Fig.5] represents an example of a floating current signal obtained in the collection step 110. .

[0055] In the derivation step 120, the floating current signal obtained in the step 110 is derived to obtain a floating current derivative signal. The graph of [Fig. 6] represents an example of a floating current derivative signal obtained in the derivation step 120 (this is the derivative of the floating current signal shown in the graph of [Fig. 5]).

[0056] In step 130, the floating current derivative signal is decomposed according to a decomposition into empirical modes. It should be noted that the floating current signal is generally too “smooth” and that it is difficult to decompose into empirical modes, which is why we are interested in its derivative, even if nothing could suggest at first glance that this floating current derivative signal could contain relevant information on the state of health of the cell.

[0057] Empirical mode decomposition consists of decomposing a signal in the form of a sum of functions, in a similar way to what Fourier series decomposition or wavelet decomposition does.

[0058] One of the particularities of the decomposition into empirical modes is that the basis of functions into which the signal is decomposed is not known a priori, but it is constructed adaptively according to the properties of the signal.

[0059] The empirical mode decomposition corresponds to the first part of the Hilbert-Huang transform (HHT). The empirical mode decomposition consists of decomposing a signal in the form of a sum of a residual signal and intrinsic mode functions (IMF for Intrinsic Mode Function). In the present application, these intrinsic mode functions are also called “intrinsic components”.

[0060] As previously indicated, the intrinsic components are not defined analytically. Rather, they are determined adaptively based on the properties of the signal.

[0061] An intrinsic component (IMF) resulting from an empirical mode decomposition (EMD) must satisfy the following requirements: - the number of extrema (i.e. the sum of the number of local maxima and the number of local minima) and the number of zero crossings of the intrinsic component must be equal or differ by a maximum of one; - at any point of the intrinsic component, the average value of the envelope defined by the local maxima and of the envelope defined by the local minima is zero.

[0062] A signal s(t) decomposed by EMD can then be written in the form: [0°63]

[0064] In this expression, r(t) corresponds to the residual signal, N is the number of intrinsic components of the EMD decomposition, and is the intrinsic component of index i. Each successive intrinsic component A contains oscillations of lower frequency than the previous one. The residual signal corresponds to a general trend of the signal s(t).

[0065] The decomposition into empirical modes involves a succession of sifting processes. The first sifting process takes the signal s(f) directly as input. The sifting process corresponds to identifying all the local extrema of the input signal, and to connecting the local maxima, respectively the local minima, by an interpolation by cubic splines, in order to obtain an upper envelope, respectively a lower envelope. An average between the upper envelope and the lower envelope can then be calculated and subtracted from the input signal. If the intermediate signal obtained (subtraction of the input signal with the average of the upper and lower envelopes) is not an intrinsic component, the sifting process is repeated on the intermediate signal (which becomes therefore the input signal of a new sieving process) until an intrinsic component is obtained. The sieving processes are repeated until the last intrinsic component is obtained, i.e. for example until the intermediate signal becomes monotonic or has only one local extremum. The remaining signal then corresponds to the residual signal Kl).

[0066] A stopping criterion may be defined for the sieving process. For example, the stopping criterion is satisfied if the standard deviation between the results of two successive sieving processes is less than or equal to a predetermined stopping threshold. The stopping threshold may typically be between 0.2 and 0.3.

[0067] The paper "The empirical mode decomposition and the Hilbert spectrum for non-linear and non-stationary time series analysis", Norden E. Huang et al., Proc. R. Soc. Lond. A (1998) 454, pp. 903-995, describes empirical mode decomposition in detail, particularly in sections 4 and 5.

[0068] Empirical mode decomposition algorithms are available in programming libraries, for example in MATLAB or Python language.

[0069] The graph of Figure 7 represents an example of decomposition into empirical modes of the floating current derivative signal represented in Figure 6. In the example considered and illustrated in Figure 7, the decomposition into empirical modes gave a single intrinsic component 34 (61) and a residual signal 33. It should be noted that, in other examples, a larger number of intrinsic components can be obtained. The number of intrinsic components, however, generally remains less than five. It is advantageous to set the stopping threshold at a relatively low level, of the order of 0.2, to extract a maximum of information from the floating current derivative signal. Using a lower stopping threshold imposes particularly long calculation times.

[0070] The intrinsic components obtained in step 130 are then used in step 150 to determine an incidence value representative of the state of health of the cell for the CV phase considered.

[0071] Different methods can be considered to determine the incidence value associated with the CV phase considered.

[0072] According to a first example, and as illustrated in [Fig.2], the determination 150 of the incidence value may comprise: - a calculation 151 of an energy for each intrinsic component, - a calculation 152 of a total intrinsic energy equal to the sum of the energies intrinsic components, and - a determination 153 of the incidence value as a function of the total intrinsic energy thus calculated.

[0073] The energy Et of an intrinsic component ci corresponds for example to the integral of the square of the amplitude of the intrinsic component ci over the acquisition duration of the CV phase considered: [°074]

[0075] The total intrinsic energy E of the CV phase considered can then be written in the form: [°076]

[0077] It should be noted that nothing would prevent, in a variant, the calculation of the total intrinsic energy E by summing the energies of only a subset of the intrinsic components obtained by the EMD decomposition (for example by considering only a predefined maximum number of the first intrinsic components obtained by the EMD decomposition).

[0078] The incidence value of the CV phase considered can then correspond to the total intrinsic energy of the CV phase, or to an average value of the total intrinsic energy for all or part of the total intrinsic energies calculated for previous CV phases.

[0079] According to a second example, and as illustrated in [Fig.3], the determination 150 of the incidence value may comprise: - a calculation 156 of a spectral density for each intrinsic component, - a calculation 157 of a total intrinsic spectral density equal to the sum of the spectral densities of the intrinsic components (or possibly equal to the sum of the spectral densities of the intrinsic components of only a subset of the intrinsic components), and - a determination 159 of the incidence value as a function of the total intrinsic spectral density thus calculated.

[0080] For example, the incidence value of the CV phase under consideration may correspond to the area of ​​the total intrinsic spectral density of the CV phase, or to an average value of the area of ​​the total intrinsic spectral density for all or part of the total intrinsic energies calculated for previous CV phases. The incidence value could also correspond to a set of samples of the total intrinsic spectral density.

[0081] However, nothing would prevent the incidence value of a CV phase from being determined by combining the total intrinsic energy and the total intrinsic spectral density. For example, the incidence value may correspond to a pair of values ​​comprising the total intrinsic energy and the area (or a set of samples) of the total intrinsic spectral density of the CV phase considered.

[0082] The spectral densities of the different intrinsic components can in particular be calculated by the Welch method. However, nothing would prevent the use of other methods of spectral density estimation, such as the Bartlett or Blackman-Tukey methods.

[0083] The spectral density of each intrinsic component can optionally be calculated from a Hilbert transform of the intrinsic component. For this purpose, and as illustrated in [Fig. 3], the step 150 of determining the incidence value comprises an additional step 155 of calculating a Hilbert transform of each intrinsic component of the floating current derivative signal, prior to the step 156 of calculating the spectral densities of the intrinsic components.

[0084] The Hilbert transform allows a real signal to be extended into the complex domain. The transformed signal then exhibits zero amplitude responses at zero frequencies. This avoids artifacts when exploiting the information contained in the processed signal. The Hilbert transform thus optimizes the spectral density calculation.

[0085] Algorithms allowing these Hilbert transform calculations and spectral density estimation are available in programming libraries, for example in MATLAB or in Python.

[0086] As illustrated in [Fig.l], the method 100 according to the invention may also comprise an optional step 140 of estimating the statistical reliability of the CV phase considered, and filtering the CV phase if it is deemed unreliable.

[0087] This filtering of a CV phase deemed unreliable makes it possible to avoid taking into account aberrant values ​​in the analysis of the health status of the cell.

[0088] The statistical reliability of a CV phase is estimated based on the intrinsic components of the floating current derivative signal obtained for the CV phase.

[0089] According to a first example, the statistical reliability of the CV phase is estimated as a function of an entropy calculated for the sum of the intrinsic components of the floating current derivative signal (for example for the sum of all the intrinsic components obtained by the EMD decomposition, or for the sum of a predefined maximum number of the first intrinsic components obtained by the EMD decomposition). Different entropy calculation methods can be envisaged, such as for example a Shannon entropy calculation, or a Kolmogorov entropy calculation. For example, the CV phases for which the calculated entropy is too low (less than a predetermined entropy threshold) are filtered. A Shannon entropy threshold of between 0.25 and 0.5 can in particular be envisaged.

[0090] According to a second example, the statistical reliability of the CV phase is estimated by comparing the total intrinsic energy calculated for the CV phase with a predetermined energy threshold. For example, CV phases that have an aberrant total intrinsic energy value (greater than the energy threshold) are filtered.

[0091] According to yet another example, the statistical reliability of the CV phase is estimated by comparing the total intrinsic energy calculated for the CV phase with the total intrinsic energies calculated for all or part of previous CV phases (for example, the total intrinsic energies can be compared with each other, or the total intrinsic energy of the current CV phase can be compared with an average value of total intrinsic energies of previous CV phases). Different statistical tests can be envisaged for this purpose (Pierce test, Pierson test, etc.).

[0092] Different criteria can be used, individually or in combination, to detect a cell failure from the incidence value determined for one or more CV phases.

[0093] According to a first example, it is possible to compare the incidence value with a predetermined incidence threshold, for one or more CV phases. If the incidence value (for example the total intrinsic energy, or the area of ​​the total intrinsic spectral density) is greater than the incidence threshold, then an alert can be raised. A failure can be detected when a certain number of alerts is reached.

[0094] According to a second example, for each new CV phase considered it is possible to carry out a linear interpolation on the incidence values ​​determined respectively for the new CV phase and for each of a certain number of previous CV phases, and to compare the slope of the linear interpolation with a predetermined slope threshold. If the slope is greater than the slope threshold (for example if the slope is strictly positive), then an alert can be raised. A failure can be detected when a certain number of alerts is reached.

[0095] According to a third example, it is possible to compare the incidence value of a CV phase with an incidence value of a previous CV phase. For example, it is possible to verify whether, for a set of several incidence values ​​corresponding to several previous CV phases, the incidence value of the most recent CV phase is greater than or equal to the incidence value of the oldest CV phase.

[0096] Figures 8 to 12 illustrate, by way of non-limiting example, how these different criteria can be combined to detect a cell failure. In the example considered, and as illustrated in [Fig. 8], the evaluation step 160 of the criterion for detecting a cell failure initially comprises a check 161 whether the incidence value is greater than an incidence threshold. If this is the case, then a check 162 is carried out to see if the slope of the linear interpolation of the incidence values ​​determined for the last seven CV phases considered is strictly positive. If this is the case, then a check 163 is carried out to see if, among the last seven CV phases considered, the most recent incidence value is strictly greater than the oldest incidence value. If this is the case, then an alert is raised. As soon as at least one of the conditions verified in steps 161 to 163 is not satisfied, no alert is raised. When three alerts are raised for three CV phases considered consecutively, then a failure is detected.

[0097] In the example considered and illustrated in Figures 9 to 12, an incidence value for a CV phase corresponds to a sliding average value of the total intrinsic energies calculated for the last twenty CV phases considered (including the current CV phase). The incidence threshold is set at 0.15 A2s '.

[0098] It should be noted that many other variants would be possible. For example, it would be possible to consider that the incidence value of a CV phase corresponds to the total intrinsic energy calculated for said CV phase (without using a sliding average), and to perform the linear interpolation on a larger number of intrinsic values ​​(for example fifty). It would also be possible to use the area of ​​the total intrinsic spectral density as the incidence value, instead of the total intrinsic energy. Also, other values ​​could be considered for the incidence threshold, the slope threshold, the number of values ​​to be considered for the linear interpolation, the number of values ​​to be considered for the sliding average, the number of alerts to be raised to trigger the detection of a failure, etc.These particular choices for the implementation of the evaluation step 160 of the criterion for detecting a cell failure are only variants of the invention.

[0099] In the example illustrated in [Fig.9], the current incidence value (incidence value of index seven) is less than the incidence threshold, and the slope of the linear interpolation of the last seven incidence values ​​is negative (in other words, the conditions of checks 161 and 162 are not satisfied). Therefore, no alert is raised. Although the previous incidence values ​​of index one, two, three, five and six were greater than the incidence threshold, no alert had been raised either because the slope of the linear interpolation of the last seven incidence values ​​was also negative at these times.

[0100] In the example illustrated in [Fig. 10], the current incidence value (incidence value of index seven) is greater than the incidence threshold, and the slope of the linear interpolation of the last seven incidence values ​​is strictly positive (the conditions of checks 161 and 162 are therefore satisfied). However, no alert is raised because the current incidence value is less than the incidence value of index one (the condition of check 163 is therefore not satisfied).

[0101] In the examples illustrated in Figures 11 and 12, the current incidence value (incidence value of index seven) is greater than the incidence threshold, the slope of the linear interpolation of the last seven incidence values ​​is strictly positive, and the current incidence value is greater than the incidence value of index one (the conditions of checks 161 to 163 are therefore all satisfied and an alert is raised. When three alerts are raised for three consecutive CV phases, then a failure is detected.

[0102] It may be advantageous to know whether a detected failure is linked to the environment in which the cell has evolved. If the failure is linked to the environment, it is possible to take corrective action at the level of the cell's environment, and the failure will potentially have a limited impact over time. A failure that is not linked to the environment is potentially more serious since it predicts an intrinsic failure of the cell, such as for example a failure linked to a manufacturing defect or to early degradation of the cell. An intrinsic failure of the cell may require the cell to be made safe to avoid a risk of accident (swelling of the cell, risk of rupture of a casing enclosing the cell, thermal runaway, risk of fire or explosion, etc.).

[0103] For this purpose, and as illustrated in [Fig. 1], the method 100 may include an optional verification step 180 if a detected failure is linked to the environment in which the cell has evolved.

[0104] This verification 180 may in particular comprise a comparison, for a given period, of the incidence values ​​stored for the cell during said period with incidence values ​​determined and stored for at least one other cell subjected to the same environment during said period. If the incidence values ​​observed for one or more other cells subjected to the same environment highlight a similar abnormal evolution during a given period, then it is highly probable that this abnormal behavior is linked to the environment (for example because of an exceptional increase in the temperature experienced by the different cells during this period).

[0105] In a variant, the verification 180 may comprise a comparison, for a given period, of environmental measurements carried out and stored during said period with a predetermined threshold. These may be, for example, measurements of the temperature experienced by the cell.

[0106] By way of example, figures 13 and 14 illustrate the evolution over time of the incidence value of a battery cell (curve 35) as well as the evolution over time, during the same period, of the temperature experienced by the cell (curve 36).

[0107] In the example illustrated in [Fig. 13], the failure highlighted by the peak in incidence value is most likely related to the environment because a similar peak is observed at the same time for the temperature experienced by the cell.

[0108] On the other hand, in the example illustrated in [Fig.14], there is no particular correlation between the change in the incidence value and the change in temperature. The failure highlighted by the sudden increase in the incidence values ​​therefore does not appear to be linked to the environment.

[0109] [Fig. 15] schematically represents a device 10 for detecting a de failure of a cell 21 of a Lithium-Ion battery 22. The device 10 comprises in particular a memory 11, a battery management system 13 and a computing unit 12 connected to the memory 11 and to the battery management system 13.

[0110] The battery management system is configured to provide current measurements made at the cell 21 during a CV phase of a CC-CV charge cycle.

[0111] The calculation unit 12 is configured to implement the method 100 according to any one of the implementation modes described above.

[0112] The device 10 may further comprise a sensor configured to measure the environment (for example a temperature sensor).

[0113] The battery management system 13 may also be configured to provide floating current measurements of one or more other cells 22 subjected to the same environment as the cell 21.

[0114] The above description clearly illustrates that, through its various features and their advantages, the present invention achieves the set objectives. In particular, monitoring the total intrinsic energy or the total intrinsic spectral density of the cell effectively allows for early detection of a failure. In laboratory tests, it was possible to detect several days in advance, or even several weeks in advance, the failure of a battery cell (the failure is detected early before a swelling of the cell which generally leads to a rupture of the cell casing).

Claims

Claims

1. Method (100) for detecting a failure of a cell (21) of a Lithium-Ion battery (20), the method (100) comprising, for at least one “constant voltage” phase, or CV phase (32), of a “constant current - constant voltage” charging cycle, or CC-CV charging cycle, of the cell: - a collection (110) of a plurality of current measurements carried out at the cell (21) during said at least one CV phase, said plurality of measurements forming a “floating current” signal, - a derivation (120) of the floating current signal to obtain a floating current derivative signal, - a decomposition into empirical modes (130) of the floating current derivative signal in order to obtain a representation thereof in the form of a sum of a residual signal and one or more intrinsic components, - a determination (150) of an incidence value representative of a cell health status (21),for said at least one CV phase, from the intrinsic components thus obtained, - an evaluation (160) of a criterion for detecting a failure of the cell (21) as a function of the incidence value of said at least one CV phase.,

2. Method (100) according to claim 1, wherein the determination (150) of the incidence value, for said at least one CV phase, comprises: - a calculation (151) of an energy for each intrinsic component, - a calculation (152) of a total intrinsic energy equal to a sum of the energies of the intrinsic components, - a determination (153) of the incidence value as a function of the total intrinsic energy thus calculated.

3. Method (100) according to any one of claims 1 to 2, wherein the determination (150) of the incidence value, for said at less one CV phase, comprises: - a calculation (156) of a spectral density for each intrinsic component, - a calculation (157) of a total intrinsic spectral density equal to a sum of the spectral densities of the different intrinsic components, - a determination (159) of the incidence value as a function of the total intrinsic spectral density thus calculated.

4. The method (100) of claim 3, wherein, for each intrinsic component, the spectral density of the intrinsic component is calculated from a Hilbert transform of the intrinsic component.

5. Method (100) according to any one of claims 1 to 4, further comprising, for said at least one CV phase, an estimation (140) of a statistical reliability of the CV phase, as a function of the intrinsic components of the floating current derivative signal, and a filtering of the CV phase if it is judged to be unreliable.

6. The method (100) of claim 5, wherein the statistical reliability of the CV phase is estimated based on an entropy calculated for a sum of the intrinsic components of the floating current derivative signal.

7. A method (100) according to claim 5, in combination with claim 2, wherein the statistical reliability of the CV phase is estimated by comparing the total intrinsic energy with a predetermined energy threshold, or with the total intrinsic energies calculated for all or part of previous CV phases.

8. Method (100) according to any one of claims 1 to 7, in which the evaluation (160) of the criterion for detecting a failure of the cell (21) comprises a comparison of the incidence value with a predetermined incidence threshold, for said at least one CV phase.

9. Method (100) according to any one of claims 1 to 8, in which the evaluation (160) of the criterion for detecting a failure of the cell (21) comprises a comparison of a slope of a linear interpolation carried out for a set of several incidence values ​​corresponding to several consecutive CV phases with a predetermined slope threshold.

10. Method (100) according to any one of claims 1 to 9, wherein the evaluation (160) of the criterion for detecting a failure of the cell (21) comprises a comparison of the incidence value for said at least one CV phase with an incidence value of a previous CV phase.

11. Method (100) according to any one of claims 1 to 10 wherein, when a failure is detected, the method further comprises a verification (180) whether the detected failure is linked to the environment in which the cell (21) has evolved.

12. Method (100) according to claim 11, wherein the verification (180) whether the detected failure is linked to the environment comprises a comparison, for a given period comprising said at least one CV phase, of incidence values ​​determined for the cell (21) during said period with incidence values ​​determined for at least one other cell (22) subjected to the same environment during said period.

13. Method (100) according to any one of claims 11 to 12, wherein the verification (180) whether the detected failure is linked to the environment comprises a comparison, for a given period comprising said at least one CV phase, of environmental measurements carried out during said period with a predetermined threshold.

14. Device (10) for detecting a failure of a cell (21) of a Lithium-Ion battery (20), said device (10) comprising: - a battery management system (13) configured to provide current measurements carried out at the cell (21) during a "constant voltage" phase, or CV phase, of a "constant current - constant voltage" charging cycle, or CC-CV charging cycle, of the cell (21), - a computing unit (12) connected to the battery management system (13), said computing unit (12) being configured to implement a method according to any one of claims 1 to 13.

Images