Method for determining probe signals, method for inspecting and detecting faults in electrochemical systems using such signals - Patents.com

Abstract

One aspect of the present invention relates to a method (100) for determining a probe signal for acoustically interrogating an electrochemical system (BAT), the method comprising: a step (1E1) of transmitting a first signal, referred to as a calibration signal (SC), comprising a wave train of a plurality of frequencies, each frequency having the same associated spectral density; a step (1E2) of receiving a response (RE) of the electrochemical system (BAT) to the calibration signal (SC); a step (1E3) of determining, for each frequency of the calibration signal (SC) based on the received response (RE); a step (1E4) of determining, for each frequency of the calibration signal (SC), a correction factor in accordance with the attenuation factor, such that a correction factor is associated with each frequency of the calibration signal (SC); and a step (1E5) of determining a probe signal (SS) obtained by multiplying the spectral density associated with each frequency of the calibration signal (SC) by the correction factor corresponding to that frequency.

Description

[Technical Field]

[0001] The technical field of the invention is that of non-invasive diagnostics and operation of electrochemical systems that convert energy (cells, batteries, fuel cells, etc.).

[0002] The present invention relates to the determination of a probe signal and the use of this signal in monitoring and detecting faults in electrochemical systems. The present invention also relates to devices configured to implement such detection. [Background technology]

[0003] Systems for monitoring and managing batteries are essential to meet user demands as effectively as possible. Generally, monitoring and management systems take into account the battery's charging stage, its state of health, and / or its safety state (detection of abnormal behavior). This variety of information is generally obtained by measuring the electrical magnitude at the battery's terminals, but also by external data such as that provided by thermal analysis techniques (temperature and / or heat flux measurements) or non-destructive control (e.g., acoustic characterization). From the information obtained by measurements, the management system adjusts the battery's operation to optimize its performance and ensure the safety of the assembly.

[0004] However, currently proposed techniques for monitoring operating batteries via ultrasonic acoustic signals (Gold et al., 2017; Sood, Pecht, and Osterman, 2016; Steingart et al., 2016, DE102015210266A1) have several drawbacks. First, these techniques only consider specific operational indicators of the battery (state of charge, state of health, temperature, fault detection) that are independent of each other and do not necessarily take into account all of the information contained in the measured acoustic signals.

[0005] The characteristic parameters of the incident acoustic signal used are then often selected more or less arbitrarily and do not take into account the particularities of each battery, resulting in the loss of relevant information for optimal management of the battery and the inability to take into account changes in the battery's characteristics over its service life.

[0006] Finally, the majority of works have used incident signals in the form of pulses of short duration (tens of nanoseconds to a few milliseconds), which do not allow the establishment of a steady-state regime within the battery and thus limit the power of the mathematical tools used to analyze the signal.

[0007] The choice of the frequency of the incident signal is very important: it depends substantially on the experimental assembly used (the characteristics of the transducers used to transmit the incident signal and to receive the transmitted signal, the geometric particularities of the battery to be tested, the nature of the materials that make up the battery, etc.), the information expected (measurement of the time of flight, frequency or time analysis of the transmitted signal), and the structural characteristics of the material whose changes it is desired to monitor (thin layers of electrodes, dense or porous materials, etc.). [Prior art documents] [Patent documents]

[0008] [Patent Document 1] German Patent Application Publication No. 102015210266 Summary of the Invention [Problem to be solved by the invention]

[0009] Therefore, there is a need for an effective method that allows determining the optimal parameters (frequency, duration, amplitude) of the incident acoustic signal in order to detect fluctuations and changes in the transmitted signal with maximum accuracy. [Means for solving the problem]

[0010] The present invention provides a solution to the above-mentioned problems by enabling the determination of probe signal characteristics that are specific to the electrochemical system being acoustically tested. In addition, the adaptability of the probe signal makes it possible to account for different operating conditions and changes in the electrochemical system over its lifetime, as well as to detect anomalies in the operation of the electrochemical system.

[0011] Thus, a first aspect of the present invention relates to a method for determining a probe signal for acoustically interrogating an electrochemical system, said method comprising: - transmitting a first signal, called a calibration signal, that includes a wave train of multiple frequencies, each frequency having an identical associated spectral density; - receiving a response of the electrochemical system to the calibration signal; - determining a signal attenuation factor for each frequency of the calibration signal based on the received responses; - determining, for each attenuation factor, a correction factor according to said attenuation factor, so as to associate a correction factor with each frequency of the calibration signal; - determining a probe signal, said probe signal being obtained by multiplying the spectral density associated with each frequency of the calibration signal by a correction factor corresponding to said frequency.

[0012] According to the present invention, the determined probe signal takes into account the specificity of the electrochemical system to be interrogated. In addition, the system's response to the probe signal comprises a characteristic spectral density, the changes of which are easy to observe, thus greatly improving the detection of aging or faults in the electrochemical system.

[0013] In addition to the features described in the previous paragraph, the method according to the first aspect of the invention may have one or more of the following additional features, taken individually or in any technically permissible combination:

[0014] Advantageously, the spectral density (and therefore amplitude) associated with each frequency in the calibration signal is selected to induce a linear response of the electrochemical system.

[0015] Advantageously, the duration of the calibration signal is long enough to establish a steady state regime in the response of the electrochemical system.

[0016] Advantageously, frequencies associated with a coefficient of attenuation greater than a predetermined threshold are not considered in determining the probe signal.

[0017] Advantageously, during the step of transmitting the first signal, the calibration signal is transmitted while the electrochemical system is in a reference state.

[0018] A second aspect of the present invention relates to a method for testing the conformance of an electrochemical system to be tested against a reference electrochemical system, the method comprising: - determining a probe signal using the method according to the first aspect of the invention implemented on a reference electrochemical system; For each calibration electrochemical system of the plurality of calibration electrochemical systems: - transmitting the probe signal determined above; - receiving a response of a calibration electrochemical system to the probe signal; Thus, statistics of the response of the calibration system are obtained; this method also: - transmitting a probe signal within the electrochemical system to be tested; - receiving a response of the electrochemical system to be tested to the probe signal; - comparing the response of the electrochemical system to be tested with statistics of the responses of a calibration electrochemical system so as to detect faults in the electrochemical system to be tested.

[0019] A third aspect of the present invention relates to a method for detecting a fault in an electrochemical system, said method comprising: - determining a probe signal using the method according to the first aspect of the invention; - transmitting the probe signal determined above; - receiving a response of the electrochemical system to the probe signal; - analyzing the response to detect a fault in the electrochemical system.

[0020] The steps of transmitting a probe signal, receiving a response of the electrochemical system, and analysing the response are preferably repeated at regular or variable time intervals.

[0021] In addition to the features described in the previous paragraph, the method according to the third aspect of the invention may have one or more of the following additional features, taken individually or in any technically permissible combination:

[0022] During the step of determining the probe signal, the probe signal is advantageously determined while the electrochemical system is in a reference state.

[0023] Advantageously, the steps of transmitting a probe signal, receiving a response of the electrochemical system, and analysing the response are implemented each time the electrochemical system is in the reference condition used to determine the probe signal during the step of determining the probe signal.

[0024] The steps to analyze the response are: - determining the spectral density of the response to the probe signal; - storing the spectral density of the response to the probe signal; - a substep of comparing the determined spectral density with statistical values ​​of responses, said statistical values ​​of responses being determined from spectral densities stored during a previous implementation of the step of storing spectral densities of responses to probe signals; detecting a fault in the electrochemical system when a difference between the spectral density determined during the determining substep and the response statistics is greater than a predetermined value; Advantageously, it comprises:

[0025] Advantageously, the probe signal is determined again at regular intervals or on request of the user.

[0026] Advantageously, during the step of analyzing the response, a malfunction of the electrochemical system is detected when the rate of variation of the response deviates by more than a predefined threshold, the rate of variation of the response being defined as the difference between the last two responses measured divided by the time separating these two measurements.

[0027] Advantageously, the method according to the third aspect of the invention includes the step of transmitting a signal to a management system responsible for the electrochemical system when a fault is detected.

[0028] The method according to the third aspect of the invention comprises, after the step of analysing the response to detect a fault in the electrochemical system: - once the correction factors have been stored, determining a new probe signal using the method according to the first aspect of the invention, - analyzing the changes in the correction factor to detect possible faults; Advantageously, it comprises:

[0029] A fourth aspect of the invention relates to a device comprising means configured to implement a method according to the first, second or third aspect of the invention.

[0030] A fifth aspect of the present invention relates to a computer program comprising instructions which, when executed by a computer, cause the computer to implement a method according to the first, second or third aspect of the present invention.

[0031] A sixth aspect of the present invention relates to a computer-readable storage medium having recorded thereon a computer program according to the fifth aspect of the present invention.

[0032] The invention and its various applications will be better understood upon reading the following description and examining the accompanying figures.

[0033] These figures are presented for informational purposes and are not intended to limit the invention. [Brief explanation of the drawings]

[0034] [Figure 1] 1 is a flow diagram of a method according to a first aspect of the present invention; [Figure 2] 1 is a graph showing a calibration signal and the response of an electrochemical system to the signal. [Figure 3] 1 is a graph showing the spectral density of a calibration signal and the spectral density of the system's response to such a signal. [Figure 4] (a) Spectral density of the response and (b) associated damping coefficient. [Figure 5] (a) shows the damping coefficient of the response, and (b) shows the associated correction coefficient. [Figure 6] 4(a)(b)(c)(d) are graphs illustrating the calculation principle of the probe signal using the method according to the first aspect of the present invention; [Figure 7] 3 is a flow diagram of a method according to a second aspect of the present invention; [Figure 8] 3 is a flow diagram of a method according to a third aspect of the present invention. [Figure 9] 10(a), 10(b), and 10(c) are diagrams showing an example of detecting a defect using a method according to a third aspect of the present invention. [Figure 10] 3 is a flow diagram of a particular embodiment of a method according to the third aspect of the invention; [Figure 11] 3 is a flow diagram of a particular embodiment of a method according to the third aspect of the invention; DETAILED DESCRIPTION OF THE INVENTION

[0035] These figures are presented for information purposes and do not limit the invention. In the following, the term "spectral density of a signal" refers to the power spectral density of a signal. Likewise, the term "frequency" in the context of the invention corresponds to a frequency width centered on the frequency in question, the value of which is modified by the parameters of the measurement, and more particularly by the ratio between the bandwidth of the measurement and the sampling of the selected frequency.

[0036] Probe signal determination A first aspect of the present invention, shown in FIG. 1, relates to a method 100 for determining a probe signal SS for acoustically interrogating an electrochemical system BAT. As described in detail below, the method 100 according to the first aspect of the present invention allows for determining a probe signal SS configured to generate a characteristic response RE (or an expected response RE) of an assembly comprising the electrochemical system BAT and a transducer that generates the signal. The method 100 according to the first aspect of the present invention also allows for obtaining a probe signal SS that can vary over time in order to monitor changes in the electrochemical system BAT. For simplicity, reference will be made below to the response RE of the electrochemical system BAT to designate the characteristic response RE (or the expected response RE) of the assembly comprising the electrochemical system BAT and a transducer that generates the probe signal SS.

[0037] To this end, the method 100 according to the first aspect of the present invention includes a step 1E1 of transmitting a first signal, referred to as a calibration signal SC, which includes a wave train of multiple frequencies, preferably with the same spectral density associated with each frequency. In one embodiment, the calibration signal SC is generated by a piezoelectric transducer PZ1 fixed on the electrochemical system BAT. In one embodiment, the duration of the calibration signal SC is long enough to establish a steady-state regime in the response of the electrochemical system BAT.

[0038] In one embodiment, the spectral density (and therefore the amplitude) of the calibration signal SC is selected to ensure the linearity of the response RE of the electrochemical system BAT. To check this condition, the calibration signal SC can be transmitted successively at different amplitudes, for example by adjusting the voltage of the signal transmitted to the piezoelectric transducer PZ1. For all measurements, the energy of the transmitted signal (i.e., the response RE of the electrochemical system BAT) must be proportional to the energy of the transmitted calibration signal SC.

[0039] In one embodiment, the calibration signal SC is defined by the frequency range considered (minimum and maximum frequencies), the sampling frequency (the precise frequencies at which the incident signal is constructed), the spectral density associated with each frequency, and the number of consecutive periods required at each frequency to reach a steady-state regime.

[0040] The method 100 according to the first aspect of the invention then comprises a step 1E2 of receiving a response RE of the electrochemical system BAT to the calibration signal SC, which may be performed by a transducer PZ2 (e.g. a piezoelectric transducer) fixed to the electrochemical system BAT.

[0041] The signals measured during the transmitting step 1E1 and the receiving step 1E2 are shown in FIG. 2, which shows the calibration signal SC and the response RE of the electrochemical system BAT. In this figure, it is possible to identify two relative transition regimes, the beginning and the end of the calibration signal SC, as well as a steady-state regime in the response RE of the electrochemical system BAT. Obtaining the steady-state regime in the response RE of the electrochemical system BAT is preferred for determining the probe signal SS, and therefore the calculations required to obtain the probe signal SS described below are preferably used from data acquired during the steady-state regime.

[0042] As shown in FIG. 3, it is possible to analyze the transmitting step 1E1 and the receiving step 1E2 from a frequency perspective, taking into account the spectral densities of the calibration signal SC ("transmitted") and the response of the electrochemical system BAT ("received") according to frequency. In the example shown in FIG. 3, the spectral density of the received signal (i.e., the response RE of the electrochemical system BAT) varies substantially over the measured frequency range, whereas the spectral density of the calibration signal SC is constant over most of the frequency band.

[0043] The frequency band is generally selected from previous measurements on the same system BAT and / or according to the nature and characteristics of the transducers PZ1, PZ2 used for generating the calibration signal SC (or probe signal SS) and / or measuring the response RE of the electrochemical system BAT.

[0044] From this information, it is possible to consider attenuating the received signal, i.e., the response RE of the electrochemical system BAT, by comparison with the transmitted signal, i.e., the calibration signal SC. For this purpose, the method according to the invention also includes a step 1E3 of determining a signal attenuation coefficient for each frequency of the calibration signal SC based on the received response RE. The attenuation for a given frequency can be defined as the spectral density of the received signal (i.e., the response RE of the system) at a given frequency divided by the spectral density of the transmitted signal (i.e., the calibration signal SC) at a given frequency. For a given spectral density of the calibration signal SC at the same frequency, the greater the attenuation, the smaller the spectral density of the response RE of the electrochemical system BAT. This step is illustrated in FIGS. 4a and 4b. FIG. 4a illustrates the spectral density of the response RE of the electrochemical system BAT to the calibration signal SC. As indicated by the dotted line in this figure, the calibration signal SC is characterized by a substantially constant spectral density across the frequency band. [Figure 4b] illustrates the attenuation coefficient associated with each frequency; the further the spectral density of that frequency is from the dotted line, the higher its attenuation coefficient.

[0045] From these attenuation coefficients, correction coefficients can then be determined, which are then applied to the calibration signal SC to obtain the probe signal SS. For this reason, the method 100 according to the first aspect of the invention comprises a step 1E4 of determining, for each attenuation coefficient, a correction coefficient according to said attenuation coefficient, so as to associate a correction coefficient with each frequency of the calibration signal SC. This step is illustrated in Figures 5a and 5b. As can be seen in these two figures, a high attenuation coefficient (compared to the others) is associated with a high correction coefficient (compared to the others). This relationship is understood in that the correction coefficients attempt to offset the effect of attenuation.

[0046] As mentioned above, the correction factors are then used to determine the probe signal SS from the calibration signal SC, which has made it possible to determine these correction factors. For this reason, the method according to the first aspect of the invention comprises a step 1E5 of determining the probe signal SS, which is obtained by multiplying the spectral density associated with each frequency of the calibration signal SC by the correction factor corresponding to that frequency. [Figure 6] shows in [Figure 6a] a representation of the spectral density of the calibration signal SC and in [Figure 6b] the correction factors determined in the preceding step that make it possible to obtain the response RE shown in [Figure 6d], i.e., a response RE with a spectral density of substantially constant power. The spectral density of the probe signal SS shown in [Figure 6c] that makes it possible to obtain this response RE is obtained by multiplying the spectral density of the calibration signal SC associated with each frequency by the corresponding correction factor.

[0047] In one embodiment, the correction coefficients associated with each frequency (already implemented as shown in FIG. 6b) are selected so that, upon transmission of the probe signal SS, the spectral density associated with each frequency in the response RE of the electrochemical system BAT is substantially identical for each frequency (already implemented as shown in FIG. 6d). Having the response RE to the probe signal SS with a substantially constant spectral density across the entire frequency range makes it very easy to identify, with high precision, fluctuations observed during operation of the electrochemical system BAT. The fluctuations in the energy of the transmitted signal can then be quantified relative to a reference value, and the frequency at which the relative fluctuations are greatest can be identified. This comparison capability is further enhanced when the probe signal SS is acquired relative to a reference condition and a comparison is made between two measurements performed relative to this same reference condition. In an alternative embodiment, the correction coefficients associated with each frequency are selected so that the spectral density associated with each frequency in the response RE of the electrochemical system BAT to transmission of the probe signal SS increases or decreases with frequency. In an alternative embodiment, the correction coefficients associated with each frequency are selected so that the spectral density associated with each frequency in the response RE of the electrochemical system BAT to transmission of the probe signal SS is a parabolic function of that frequency.

[0048] In one embodiment, frequencies associated with damping coefficients greater than a predetermined threshold are not considered, e.g., only frequencies of the response RE of the electrochemical system BAT that have a power spectral density greater than 30 dB / Hz are retained.

[0049] It is advantageous to determine the probe signal SS with respect to a well-defined state of the electrochemical system BAT, referred to as the reference state. As will be seen below, such a reference state can be advantageous for monitoring changes in the electrochemical system BAT over time. Thus, in one embodiment, during the step 1E1 of transmitting a first signal, the calibration signal SC is transmitted while the electrochemical system BAT is in the reference state. For example, when the electrochemical system BAT is a battery, the reference state can correspond to a given state of charge. In one embodiment, this reference state can be associated with some operating condition of the electrochemical system BAT. For example, when the electrochemical system BAT is a battery, the reference state can correspond to a given state of charge of the battery and a given temperature, or a given state of charge of the battery, a given temperature, and a given supply current.

[0050] Quality control of electrochemical systems From the probe signal SS determined using the method 100 according to the first aspect of the present invention, it is then possible to characterize the state of the electrochemical system BAT. A first possible application is the quality control of electrochemical systems BAT in a production line, which produces electrochemical systems BAT suitable for testing at the end of the production chain. For this purpose, each electrochemical system BAT at the end of production is compared with a statistical value of the response RE established over a number of calibrated electrochemical systems BAT, this statistical value having been obtained using a probe signal SS determined from a reference electrochemical system BAT corresponding to the desired product. This comparison or quality control can therefore be carried out by utilizing a probe signal SS such as that determined by the method 100 according to the first aspect of the present invention.

[0051] For this reason, a second aspect of the invention, shown in FIG. 7, relates to a method 200 for testing the compatibility of an electrochemical system BAT to be tested against a reference electrochemical system BAT.

[0052] The method 200 according to the second aspect of the invention comprises a step 2E1 of determining a probe signal SS using the method 100 according to the first aspect of the invention, said method 100 being implemented on a reference electrochemical system BAT. This step makes it possible to obtain a probe signal SS corresponding to a target quality on the production line.

[0053] After the probe signal SS has been determined, it is necessary to establish statistics of the responses RE to this probe signal SS. To this end, the method 200 according to the second aspect of the invention comprises a step 2E2 of transmitting the probe signal SS determined above to each calibration electrochemical system BAT of the plurality of calibration electrochemical systems BAT; and a step 2E3 of receiving the responses RE of the calibration electrochemical system BAT to said probe signal SS. Thus, statistics of the responses RE to the probe signal SS are constructed, which can then be used to test the electrochemical systems BAT leaving the production chain.

[0054] For this purpose, the method 200 according to the second aspect of the present invention includes a step 2E4 of transmitting a probe signal SS in the electrochemical system BAT to be tested; a step 2E5 of receiving a response RE of the electrochemical system BAT to be tested to said probe signal SS; and a step 2E6 of comparing the response RE of the electrochemical system BAT to be tested with statistical values ​​of the responses of a calibration electrochemical system BAT so as to detect faults in the electrochemical system BAT to be tested.

[0055] In one embodiment, the electrochemical system BAT involved in the method 200 according to the second aspect of the present invention is in the same reference state when the probe signal SS is determined or when subjected to the probe signal SS so determined.

[0056] Monitoring the electrochemical system BAT during operation The probe signal SS determined by the method 100 according to the first aspect of the invention can also be used to monitor the electrochemical system BAT during its operation. For this reason, a third aspect of the invention, shown in Figure 8, relates to a method 300 for detecting faults in the electrochemical system BAT.

[0057] The method 300 according to the third aspect of the invention comprises a step 3E1 of determining the probe signal SS using the method 100 according to the first aspect of the invention. In other words, the method 100 according to the first aspect of the invention is implemented on the electrochemical system BAT for which it is desired to provide monitoring.

[0058] The method 300 according to the third aspect of the invention then comprises a step 3E2 of transmitting a probe signal SS so as to acquire the acoustic response of the electrochemical system BAT.

[0059] The method 300 according to the third aspect of the invention then comprises a step 3E3 of receiving the response of the electrochemical system BAT to said probe signal SS. It is interesting to note here that the expected response is known, since the probe signal SS was obtained using the method 100 according to the first aspect of the invention. It is therefore very easy to check that the response obtained during this step matches the expected response.

[0060] For this reason, the method according to the third aspect of the invention then comprises a step 3E4 of analysing the response so as to detect possible faults in the electrochemical system.

[0061] Preferably, to ensure monitoring during operation of the electrochemical system BAT, the step 3E2 of transmitting the probe signal SS, the step 3E3 of receiving the response of the electrochemical system BAT, and the step 3E4 of analyzing the response are repeated at regular or variable time intervals.

[0062] In one embodiment, the time interval between two implementations of these steps is variable, with the time step associated with an electrochemical system BAT that is in use (high current, high temperature, etc.) being smaller than the time step associated with an electrochemical system BAT that is not in use (zero current, etc.).

[0063] As described in more detail below, the analysis of the response RE of the electrochemical system BAT to the probe signal SS can be performed in several ways. This analysis can, for example, consist of a frequency-by-frequency comparison of the expected response and the response measured during step 3E3 of receiving the response of the electrochemical system BAT. Such an example of detecting a fault from the measurement of the system's response RE is shown in FIG. 9. FIG. 9a shows the expected response RE to the probe signal SS, as well as a tolerance that defines the relative difference from the expected response RE that can be considered normal. FIG. 9b shows the response RE corresponding to a fault-free condition, and FIG. 9c shows the response RE corresponding to a faulty condition, with the portion of the response RE causing the fault indicated by a black circle. Thus, FIG. 9 illustrates how the use of the probe signal SS associated with the expected response RE of the electrochemical system BAT facilitates the determination of the tolerance.

[0064] Alternatively or additionally, this analysis may involve calculating a convolution between the spectral density of the expected response RE and the spectral density of the acquired response RE. More generally, any method of comparing the two signals may be implemented, with a too large difference between the expected response RE and the acquired response RE being evidence of a fault in the electrochemical system BAT. As shown below, this comparison may also be made from statistics of the response of the electrochemical system BAT to the probe signal SS.

[0065] It is advantageous to implement the method 300 according to the third aspect of the present invention using the probe signal SS determined for a reference state of the electrochemical system BAT for which reliable monitoring is desired. Indeed, as already detailed, using a reference state has certain advantages in terms of the ability to detect possible faults. For this reason, in one embodiment, during step 3E1 of determining the probe signal SS, the probe signal SS is determined while the electrochemical system BAT is in a reference state. Preferably, this reference state is defined for the application consisting of the electrochemical system BAT and / or for the mission profile of the electrochemical system BAT. This reference state can be selected, for example, at a partial state of charge, current, and a defined temperature range, as well as at a maximum state of charge, zero current, and a defined temperature range.

[0066] It is useful to note that selecting a reference state for determining the probe signal SS does not require that step 3E2 of transmitting the probe signal SS, step 3E3 of receiving the response of the electrochemical system BAT, and step 3E4 of analyzing the response be performed when the electrochemical system BAT is in this reference state. Conversely, these steps can also be implemented during operation between two moments when the electrochemical system BAT is in this reference state. During such implementation, the fluctuations in the energy transmitted by the probe signal SS and the fluctuations in the power spectral density observed in the response RE of the electrochemical system BAT relative to the expected response RE change reversibly and, except in the case of malfunction, follow similar changes with each cycle, approaching the reference value when the electrochemical system BAT again approaches the reference state (assuming that aging, which induces slight deviations as described below, is excluded). Therefore, using the reference state makes it possible to verify that this reversible change in the response to the probe signal SS has occurred.

[0067] Using the same alternative, it is possible to check the state of the electrochemical system BAT only when the electrochemical system BAT is in the reference state. While this will, of course, prevent the above-mentioned changes from being observed, this embodiment can be adapted to situations where it is desirable to reduce the number of steps of transmitting a probe signal SS (step 3E2), receiving a response from the electrochemical system BAT (step 3E3), and analyzing the response (step 3E4), or where changes in the response outside the reference state are of little interest. Thus, in one embodiment, the steps of transmitting a probe signal SS (step 3E2), receiving a response from the electrochemical system BAT (step 3E3), and analyzing the response (step 3E4) are implemented each time the electrochemical system BAT is in the reference state used to determine the probe signal SS. If the electrochemical system BAT is in the reference state at each step of measuring the response RE, it becomes easier to compare the measured response with the expected response.

[0068] As mentioned above, it is of interest to proceed with a statistical analysis of the response RE of the electrochemical system BAT to the probe signal SS. This approach makes it possible, in particular, to automate the determination of one or more thresholds or tolerances corresponding to malfunctions of the electrochemical system BAT. It also makes it possible to automate the updating of these thresholds by adding measurements that provide statistics of the response RE during the operation of the electrochemical system BAT. To this end, in one embodiment shown in FIG. 10, the step 3E4 of analyzing the response comprises: - a substep 3E41 of determining the spectral density of the response RE to the probe signal SS; - a substep 3E42 of storing the spectral density of the response RE to the probe signal SS; - a substep 3E43 of comparing the determined spectral density with statistical values ​​of the responses RE, said statistical values ​​being determined from the spectral densities stored during a previous implementation of the step 3E42 of storing the spectral densities of the responses RE to the probe signal SS; a substep 3E44 of detecting a fault in the electrochemical system BAT when the difference between the spectral density determined during the determining substep 3E41 and the statistical value of the response is greater than a redetermined value; Includes.

[0069] In one embodiment, a fault in the electrochemical system BAT is detected during substep 3E44 of detecting a fault in the electrochemical system when the spectral density of the response RE deviates by more than 3σ relative to the statistical value of the response for at least one frequency.

[0070] In an alternative embodiment, a fault in the electrochemical system BAT is detected when, during substep 3E44 of detecting a fault in the electrochemical system, the spectral density of the response RE is more than 3σ apart in relation to the statistical values ​​of the response for a plurality of frequencies, the number of which is selected according to the desired sensitivity. It may be useful to note that this statistical method can also be used in the framework of method 200 for checking the suitability of an electrochemical system BAT according to the second aspect of the invention, the statistical values ​​used then being obtained from a plurality of calibrated electrochemical systems BAT.

[0071] The variation of the response RE over time makes it possible to determine whether the materials of which the electrochemical system BAT is composed undergo reversible variations within predefined limits (repeatable changes with each charge-discharge cycle, temperature effects), or slow, continuous deviations (aging of the battery and device assembly), and even sudden variations exceeding predefined limits that would be a sign of malfunction. It is therefore interesting to consider these variations.

[0072] For this reason, in one embodiment, the rate of variation of the response RE of the electrochemical system BAT to the probe signal SS is also taken into account. More specifically, during step 3E4 of analyzing the response, a malfunction of the electrochemical system BAT is detected when the rate of variation of the response (defined as the difference between the last two measured responses RE divided by the time separating these two measurements) deviates by more than a predefined threshold. The predefined threshold can be modified according to the rate of variation considered normal and / or according to a statistical value of the rate of variation that is configured simultaneously with the statistical value of the response. Preferably, this verification of the variation is performed frequency by frequency. In other words, the variation of the spectral density is evaluated for each frequency so that the frequency obtains as many variation values ​​as possible. In this case, a malfunction can be observed when at least one frequency is associated with a variation greater than a predetermined threshold, or when a predefined number of frequencies are associated with a variation greater than a predefined threshold. In the second case, the number of frequencies considered can depend on the desired sensitivity of the malfunction detection. The threshold can vary according to the frequencies considered or can be the same for all frequencies.

[0073] In one embodiment, when a fault is detected, the method according to the third aspect of the invention comprises the step of sending a signal to a management system responsible for the electrochemical system so that the electrochemical system can apply corrective measures.

[0074] Monitoring the ageing of electrochemical systems It may be interesting to use the probe signal SS not only to identify faults but also to monitor the aging of the electrochemical system BAT. In fact, if the probe signal SS is determined for the same system BAT at two time-separated moments under a given reference state, the correction factor associated with the probe signal SS determined at the first moment will differ from the correction factor associated with the probe signal SS determined at the second moment for a given expected response RE. This change in the correction factor generally indicates a change in the electrochemical system BAT over time, which is induced by aging.

[0075] To take advantage of this, in one embodiment shown in FIG. 11, a method 300 according to the third aspect of the invention includes, after step 3E4 of analyzing the response to detect a fault in the electrochemical system, step 3E5 of determining a new probe signal SS using the method 100 according to the first aspect of the invention, with the correction factor stored. In addition, in this embodiment, the method 300 according to the third aspect of the invention also includes step 3E6 of analyzing changes in the correction factor to detect possible faults. In one embodiment, step 3E5 of determining a new probe signal SS is implemented only when the electrochemical system BAT is in a reference state. In one embodiment, an indicator of aging of the electrochemical system (and / or the means for generating and / or acquiring a signal) is determined from changes in the correction factor.

[0076] More generally, the changes in the correction factors can be used to characterize numerous phenomena. For example, the rate of change of these correction factors (average rate for all frequencies of the signal, maximum rate, and / or rate of one or more correction factors associated with one or more specific frequencies) can be considered to determine the rate of material degradation and even aging mechanisms of the electrochemical system BAT.

[0077] It is also possible to consider an acceleration of the change of these correction factors, which may be due to material degradation linked to strong temporary stresses of the electrochemical system BAT (e.g., operation at high power for fast charging or discharging, operation at high temperature, or charging at low temperature).

[0078] Overall, an in-depth analysis of the rate of change of the correction factor makes it possible to identify the optimal operating conditions of the electrochemical system BAT and those that should be avoided, and to objectively define the optimal operating range (range of state of charge, temperature, and maximum power) of the electrochemical system BAT. It should be noted that the optimal operating range of the electrochemical system BAT does not have to be constant over the entire service life of the electrochemical system BAT: for example, certain operating conditions that result in strong stresses can be tolerated without major problems on a new battery, but not necessarily on an end-of-life battery.

[0079] Devices implementing the present invention A fourth aspect of the present invention relates to a device comprising means configured to implement a method according to the first, second, or third aspect of the present invention. In one embodiment, the device comprises means PZ1 for generating a calibration signal SC or a probe signal SS in an electrochemical system BAT. In one embodiment, these generating means can be removably arranged on the electrochemical system. In one embodiment, the device comprises means PZ2 for receiving a response RE of the electrochemical system to the calibration signal SC or the probe signal SS. In one embodiment, these receiving means can be removably arranged on the electrochemical system. In one embodiment, the device comprises means for computing and a memory configured to store instructions required for implementing a method according to the first, second, or third aspect of the present invention. Furthermore, the computing means is configured to access the memory and execute the instructions. The computing means is configured to control the means PZ1 for generating the probe signal SS or calibrating the calibration signal SC to receive data collected by the means PZ2 for receiving the response RE of the electrochemical system BAT. In one embodiment, the device is a management system for the electrochemical system BAT, for example, a system for managing a battery.

Claims

1. A method (100) for determining a probe signal (SS) for acoustically interrogating an electrochemical system (BAT), comprising: a step (1E1) of transmitting a first signal, called a calibration signal (SC), comprising a wave train of several frequencies so as to induce a response (RE) of the electrochemical system (BAT), the spectral density associated with each frequency being identical; receiving (1E2) the response (RE) of the electrochemical system (BAT) to the calibration signal (SC); - determining (1E3) for each frequency of the calibration signal (SC) a signal attenuation coefficient based on the received response (RE); - determining (1E4), for each attenuation factor, a correction factor according to said attenuation factor, so as to associate a correction factor to each frequency of the calibration signal (SC); a step (1E5) of determining a probe signal (SS), said probe signal (SS) being obtained by multiplying the spectral density associated with each frequency of the calibration signal (SC) by a correction factor corresponding to said frequency; A method (100) comprising:

2. 2. The method (100) of claim 1, wherein the spectral density associated with each frequency in the calibration signal (SC) is selected to induce a linear response of the electrochemical system (BAT).

3. 3. A method (100) according to claim 1 or 2, characterized in that the duration of the probe signal (SC) is long enough to establish a steady state regime in the response (RE) of the electrochemical system (BAT).

4. Method (100) according to any one of claims 1 to 3, characterized in that frequencies associated with an attenuation coefficient greater than a predetermined threshold are not taken into account in determining the probe signal (SS).

5. 5. The method (100) according to any one of claims 1 to 4, characterized in that during the step (1E1) of transmitting a first signal, a calibration signal (SC) is transmitted while the electrochemical system (BAT) is in a reference state.

6. A method (200) for testing the compatibility of an electrochemical system (BAT) to be tested against a reference electrochemical system (BAT), comprising: - determining (2E1) a probe signal (SS) using the method according to any one of claims 1 to 5 implemented on a reference electrochemical system, For each calibration electrochemical system (BAT) of the plurality of calibration electrochemical systems (BAT), a step (2E2) of transmitting the probe signal (SS) determined above, receiving (2E3) a response (RE) of a calibration electrochemical system (BAT) to said probe signal (SS), Thus, statistics of the response (RE) of the calibration system are obtained, and the method also a step (2E4) of transmitting a probe signal (SS) in the electrochemical system (BAT) to be tested, receiving (2E5) the response (RE) of the electrochemical system (BAT) to be tested to said probe signal (SS); a step (2E6) of comparing the response of the electrochemical system (BAT) to be tested with the statistical values ​​of the responses of calibration electrochemical systems (BAT) so as to detect faults in the electrochemical system (BAT) to be tested; A method (200) comprising:

7. A method (300) for detecting faults in an electrochemical system, comprising: - a step (3E1) of determining a probe signal using a method according to any one of claims 1 to 5, a step (3E2) of transmitting the probe signal (SS) determined above; - receiving (3E3) the response of the electrochemical system (BAT) to said probe signal (SS); - a step (3E4) of analyzing the response so as to detect faults in the electrochemical system; Including, A method (300) characterized in that the steps of transmitting (3E2) a probe signal (SS), receiving (3E3) a response of the electrochemical system (BAT), and analyzing (3E4) the response are repeated at regular or variable time intervals.

8. 8. The method (300) of claim 7, characterized in that during the step (3E1) of determining the probe signal (SS), the probe signal (SS) is determined while the electrochemical system (BAT) is in a reference state.

9. 9. The method (300) of claim 8, wherein the steps of transmitting (3E2) the probe signal (SS), receiving (3E3) the response of the electrochemical system (BAT), and analyzing (3E4) the response are implemented each time the electrochemical system (BAT) is in the reference state used to determine the probe signal (SS) during the step (3E1) of determining the probe signal (SS).

10. The step (3E3) of analyzing the response comprises: a substep (3E41) of determining the spectral density of the response to the probe signal (SS), a substep (3E42) of storing the spectral density of the response to the probe signal (SS), a substep (3E43) of comparing the determined spectral density with a statistical value of the response (RE), said statistical value of the response (RE) being determined from the spectral densities stored during a previous implementation of the step (3E42) of storing the spectral densities of the responses (RE) to the probe signal (SS), a substep (3E44) of detecting a fault in the electrochemical system when the difference between the spectral density determined during the determining substep (3E31) and the statistical value of the response is greater than a predetermined value; 10. The method (300) according to any one of claims 7 to 9, characterized in that it comprises:

11. Method (200, 300) according to any one of claims 6 to 10, characterized in that the probe signal (SS) is redetermined at regular intervals or on request from the user.

12. A device for determining a probe signal (SS), for checking the suitability of an electrochemical system (BAT) or for detecting faults in an electrochemical system (BAT), the device comprising means configured to implement the method according to any one of claims 1 to 11.

13. A computer program comprising instructions for causing a device according to claim 12 to implement the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having the computer program according to claim 13 recorded thereon.

Images