FR3010577A1 - DEVICE AND METHOD FOR REGENERATING PILOT BATTERY BY ACOUSTIC EMISSION MEASUREMENT - Google Patents

Abstract

The invention relates to a device and a method for regenerating electrochemical energy accumulators based on acoustic emission measurement. The device comprises regeneration means comprising an electronic module for generating electrical signals connected to the accumulator, acoustic emission measuring means comprising at least one acoustic emission sensor placed on the accumulator and means for controlling the acoustic emission. the regeneration comprising means for acquiring and processing acoustic signals emitted during the generation of the electrical signals. According to the regeneration method, electrical signals are sent to the accumulator, an acoustic emission response comprising a series of acoustic events resulting from the signals is measured, and at least one acoustic emission response is determined from this acoustic emission response. regeneration characteristic of the accumulator. The invention is advantageously applied to lead accumulators.

Description

FIELD OF THE INVENTION The present invention relates to the field of the regeneration of electrochemical generators of reversible (rechargeable) energy, and in particular the regeneration of industrial and automotive accumulators, in particular lead-acid technology.

General context Electrochemical energy accumulators, whatever their type (industrial, automotive or portable), and whatever their technology (lead-acid, Li-ion, Ni-Cd, NiMH etc.) undergo over time aging phenomena resulting in particular a loss of electrical capacity and an increase in the internal electrical resistance. Several mechanisms can cause this aging, some of which may be reversible. In the case of lead acid batteries, a phenomenon leading to aging of the battery and called "sulfation" has been recognized as a reversible phenomenon. There are three different definitions for the term sulfation (Henry A. Cathelino, Fred F. Feres, Francisco Trinidad, "Sulfation in lead-acid batteries." Journal of Power Sources, 2004: 129, 113-120): 1) Sulfation refers to the general cause that leads to the death of the lead battery. It is identified experimentally by the observation of a loss of capacitance or electrical tension, an increase of the internal resistance or a decrease of the concentration of acid. 2) Sulfation evokes the formation of lead sulphate crystals in a form that is no longer electroactive, that is to say which no longer participates in the charge-discharge process. 3) Sulfation is the chemical process leading to the formation of lead sulphate. The third definition refers to the fundamental electrochemical process occurring at the electrodes during a normal discharge of the battery. Indeed, the formation of lead sulfate is essential for the operation of the accumulator. According to the other two definitions, sulfation is a factor that limits the life of the battery. The first assertion being only a phenomenological definition, the term sulfation will refer to the second definition in the rest of the present description, which corresponds to "irreversible sulfation" during normal operation. This sulphation corresponds to a reorganization of the lead sulphate crystals (PbSO4) formed during the discharge through a dissolution-recrystallization process. This mechanism leads to the formation of lead sulphate aggregates that are difficult to completely convert into active material during charging. As a result, there is less active material involved in the battery charging-discharging process, which results in a decrease in the capacity of the system. In addition, the porosity of the PbSO4 film decreases (resulting in a drop in the active surface and a less efficient material transport), which leads to an increase in the internal resistance. Such a phenomenon of sulfation can for example result from misuse (ex: too long discharge time, underload, etc.) or poor battery maintenance. This phenomenon can also be accelerated by the stratification of the electrolyte.

It is possible to extend the life of lead-acid batteries by combating this phenomenon of sulphation. For this, we implement regeneration processes (or "desulfation") that allow the recovery of the initial capacity of the battery aged lead. At present, three types of desulfation processes are known. These are electrical, chemical, or combination processes of electrical and chemical desulfation. Electrical desulfation processes consist in applying to the battery short-term electrical pulses, generally of high intensity, for a variable duration. Chemical desulfation processes involve the use of a chemical additive injected into the electrolyte, the amount of which is a function of the parameters of the battery, mainly its capacity. Distilled water may also be added to adjust the level of the electrolyte. At the end of the desulfation, the battery is fully charged in order to activate the additive and allow reuse of the battery. Finally, there are desulfation processes combining the electrical method and the chemical method. The regeneration is in this case carried out in two, or even three times. Firstly a chemical additive is injected into the battery after opening the plugs thereof. Distilled water and / or electrolyte can then be added if necessary. In a second step, the battery is connected to a desulfation machine that applies short electrical pulses to the battery. It can be implemented in a third time an electronic box that constantly sends low intensity electrical pulses.

Regeneration mechanisms are not yet well explained, and the desulfation techniques used have several disadvantages. A major disadvantage is that it is necessary to preselect the regenerable batteries, that is to say those having undergone a reversible aging mechanism, this step being very empirical and highly subjective. It relies for example on visual observations, measurements of voltage, capacity, impedance. Visual observations can detect mechanical or electrical defects that would be a sign of irreversible degradation (eg signs of leakage, corrosion, damaged terminals, etc.). Electrical and electrolyte tests (electrolyte level and density test) generally complement the visual observation of the battery.

Thus, some of the accumulators that could be regenerable or partially regenerable are generally removed for safety during the application of current techniques. However, this type of diagnosis is generally complex and time consuming to implement. Another major disadvantage of current desulphatation techniques is that the necessary regeneration time is generally not known. Regeneration according to an electric regeneration method may for example take a few hours or several days. Thus, the effective regeneration time is often established empirically and may lead to a regeneration that is not optimal (the battery is not completely regenerated), or even present risks of deterioration or even explosion or harmful (release of gas) from the battery if the regeneration time is for example too important. In addition, the uncertainty about the duration of the regeneration is a brake for a good management of the treatment of the batteries in industrial context. Currently, lead batteries are the only ones to undergo a phenomenon of degradation by sulfation which could be made reversible by a regeneration method. However, batteries based on other technologies (NiMH, NiCd, Li-ion, etc.) undergo many different degradation mechanisms, and could in the future be regenerated by similar regeneration methods, since at least one of the identified degradation phenomena can be made reversible. There is therefore a need for a fast and reliable diagnostic step for battery regeneration, to determine whether a battery is regenerable, as well as a need to know the end of the regeneration process. SUMMARY OF THE INVENTION The present invention aims to overcome the disadvantages and to meet, at least in part, the needs of the state of the art described above. In particular, the present invention aims at providing a device and a method for controlling the regeneration of electrochemical energy accumulators. The control may include a system for rapidly diagnosing whether the battery is regenerable, a system for monitoring the evolution of the regeneration and determining when the regeneration can be considered complete, also means for early diagnosis of the state of the regeneration. the battery to consider battery maintenance during its lifetime. Thus, to achieve at least one of the above-mentioned objects, among others, the present invention relates to a device for regenerating an electrochemical energy accumulator, comprising: - regeneration means comprising an electronic generator unit; electrical signals connected to said accumulator; acoustic emission measuring means comprising at least one acoustic emission sensor placed on said accumulator; - Regeneration control means comprising means for acquiring and processing acoustic signals emitted during the generation of electrical signals. The control means may comprise a storage memory of a library of characteristic acoustic signatures, means for determining regeneration characteristics of the accumulator, means for analyzing an internal state of the accumulator such as its operating state.

According to one embodiment of the invention, the battery is leaded. The control means may also include an electrochemical impedance measurement system. The present invention also relates to a method of regenerating an electrochemical energy accumulator in which: - electrical signals are sent into the accumulator; an acoustic emission response is measured comprising a series of acoustic events resulting from said signals; from said acoustic emission response at least one regeneration characteristic of the accumulator is determined. According to one implementation of the method, it is determined whether the accumulator can be at least partially regenerated. According to another implementation of the method according to the invention, the evolution of the regeneration of the accumulator by acoustic emission measurement is monitored. Preferably, a pre-established characteristic acoustic signature library is used to determine at least one regeneration characteristic of said accumulator. According to one implementation of the method, acoustic events are processed by grouping them by signal populations having characteristic common parameters. Preferably, these signal populations are compared with characteristic acoustic signatures grouped together in a pre-established acoustic signature library, different populations of acoustic events characteristic of different deterioration phenomena experienced by the accumulator at the end of the test are identified. comparing step, and determining the respective share of each of said phenomena by comparing the cumulative acoustic activity of each of the different populations of events identified. According to one implementation of the method, the electrical signals sent to said accumulator at least partially regenerate the accumulator. The library of characteristic acoustic signatures can be established prior to sending electrical signals to the accumulator for measuring an acoustic emission response, according to the following steps: a batch of electrochemical accumulators is selected; energy of the same type as the electrochemical accumulator of energy to be regenerated, each of the accumulators of the batch being in a given state of degradation; for different states of degradation of at least one accumulator of the batch, electrical signals are sent into said accumulator of the batch, at least one acoustic emission response comprising a series of acoustic signals resulting from said electrical signals is measured, and said acoustic signals; a processing of the recorded acoustic signals is carried out by grouping them by signal populations having common parameters; and determining at least one acoustic signature characteristic of a given degradation state by calibrating a relationship between at least one of the parameters common to at least one population of acoustic signals and a given degradation state of said accumulator of the batch, by means of an analysis of the values of one or more common parameters of the acoustic signal populations obtained for each of said degradation states. Other objects and advantages of the invention will appear on reading the following description of examples of particular embodiments of the invention, given by way of non-limiting examples, the description being made with reference to the appended figures 20 described. below. BRIEF DESCRIPTION OF THE FIGURES FIG. 1 is a diagram illustrating generally the regeneration device according to the invention. FIG. 2 represents an acoustic emission burst and the acoustic emission parameters (EA) calculated on this burst: amplitude (A), rise time (tm), duration (dm) between two exceedances (D1 and Dn) threshold (L), number of strokes (c). FIG. 3 is a flowchart illustrating the general principle of the implementation of the regeneration device according to the invention. FIG. 4 is a diagram showing an example of placement of acoustic emission sensors placed on a lead battery comprising 6 elements. FIGS. 5A and 5B are graphs representing the acoustic events recorded respectively on each of the two elements A and B of the same lead battery subjected to an electric load. The events are represented according to their duration of burst D, in microseconds (in abscissa) and their energy E in attojoule (in ordinate). FIG. 6 is a flowchart illustrating a method of establishing the acoustic signature library used in the diagnostic phase of the regeneration method according to the invention. FIG. 7A represents the normed linear projection on a (2D) plane of acoustic data according to the normalized eigenvectors VO and V1. FIG. 7B shows the same acoustic data according to the burst duration parameters D (in abscissa) and energy E (in the ordinate). Figure 8A shows the cumulative AC acoustic activity (burst number) of the population of pop () acoustic events during a regeneration test. FIG. 8B shows the cumulative acoustic activity of the acoustic event population papi during a regeneration test. In the figures, the same references designate identical or similar elements. DETAILED DESCRIPTION OF THE INVENTION The present invention proposes a device and a method for the regeneration of an electrochemical energy accumulator comprising a regeneration system of an accumulator, for example of electric, chemical or mixed electrical and chemical type. combined with acoustic emission measuring means for controlling the regeneration. The control means are particularly suitable for diagnosing whether the accumulator is regenerable, for monitoring the regeneration, and in particular for determining the end thereof. The present invention is particularly applicable to the regeneration of industrial and automotive accumulators, in particular lead-acid technology. Terminology An electrochemical generator is a system that converts the chemical energy contained in the active ingredients that compose it directly into electrical energy via oxidation-reduction reactions (redox reactions).

In the context of electrochemical generators, an element is a unit electrochemical cell containing two electrodes immersed in an electrolyte. It operates spontaneously in the generator direction when its electrodes are brought into contact by an external electrical circuit.

A battery is an electrochemical generator operating only in the direction of discharge. His redox reactions are irreversible. The present invention does not relate to this category of generators. An accumulator is a rechargeable electrochemical generator. Its redox reactions are reversible by connecting a power supply to its terminals creating a reverse current in the discharge direction. A battery traditionally refers to a "storage battery" connected in series and / or in parallel. The term may in particular refer to the complete system or pack comprising the elements, the integration devices, the management devices, etc. In the present description, the terms batteries and accumulators are used interchangeably to designate reversible energy electrochemical generators, whether they are unitary (element) or in the form of a set of connected elements, such as a pack. It is therefore here electrically rechargeable systems in the broad sense. By regeneration of an electrochemical accumulator is meant in the present description, the set of physico-chemical phenomena to improve the lost performance of a system by bringing them closer to its initial performance, that is to say an increase in the system capacity tending to nominal capacity and / or a decrease in the internal electrical resistance tending towards the initial resistance at the factory outlet. Regeneration within the meaning of the invention aims to extend the duration of use of electrochemical energy accumulators. In general, an accumulator is considered to be in the end of use when it no longer meets the needs of the user. For example, for a lead-acid battery, the manufacturers consider that a battery is at the end of use when its capacity reaches 80% of the nominal capacity. However this value strongly depends on the application using the battery. Regeneration according to the invention therefore aims to restore all or part of the capacity of the accumulator so that it can meet the needs of the user again, for example by restoring its capacity to a value close to its nominal value. .

In the present description, the term "desulphatation" will be used to designate the regeneration of lead-acid battery accumulators, in which reversible aging phenomena related to a sulphation process occur and are reversible, allowing regeneration of these accumulators. using specific processes and devices. The following three types of accumulators are classically distinguished: Automobile accumulators, which are batteries or accumulators intended to power a starting, lighting or ignition system. Industrial accumulators which are batteries or accumulators designed for exclusively industrial or professional purposes or used in all types of electric vehicles. Portable batteries that are sealed batteries or accumulators that can be carried by hand and do not belong to the other two categories. The present invention is particularly applicable to automotive and industrial type accumulators, including accumulators of lead-acid technology, which is the most widespread technology currently on the accumulator market. The present invention is however not limited to these two categories of accumulators, and could be adapted to the regeneration of portable accumulators, as well as battery technologies other than lead-acid technology, such as NiMH technologies, nickel-cadmium or lithium-ion. These types of accumulators may to some extent have suitable processes to improve their operation (charging / discharging) over time. These processes are therefore similar to the process of desulfating Pb batteries, and their means of implementation can be combined with the control means as in the present invention. For example, a common aging phenomenon in the accumulators consists of the formation of a deposit on the surface of the electrodes, resulting from the degradation of constituents such as the electrolyte on the surface of the active material. This deposit generally leads to an increase in the internal resistance of the systems. If a method such as the application of a current profile can destroy this layer, then it can be a regeneration process to reduce the resistance of the system.

The present invention is applicable to the extent that this process generates identifiable acoustic phenomena. Regeneration device The regeneration device according to the invention comprises regeneration control means controlled by acoustic emission measurements. With reference to FIG. 1, the regeneration device 10 according to the invention comprises regeneration means 11 connected to the battery 1 to be regenerated by connection cables 12, acoustic emission measurement means comprising emission sensors acoustic 13, and regeneration control means 14 comprising means for acquiring and processing the acoustic signals obtained by the acoustic emission measurement means 13.

The regeneration means 11 on the one hand to apply electrical profiles to the battery 1 to diagnose whether the battery is regenerable, and on the other hand to proceed to the regeneration of the battery as such. The regeneration means 11 comprise at least one electronic unit producing said electrical profiles current and / or voltage, and may also include a user interface. The regeneration means 11 of the battery may further comprise means for a chemical or mixed electrical and chemical regeneration of the battery 1. In the case of an electric or mixed electrical / chemical regeneration, the regeneration means 11 may apply electrical profiles to battery 1 at a time to determine if the battery is regenerable during an initial diagnostic phase, but also to regenerate the battery if necessary. Whatever the regeneration technique used, the electronic unit sends the battery 1 an electrical signal to allow the realization of an acoustic emission measurement, for example during an initial diagnostic phase for the control of the evolution regeneration.

The connection cables 12 make it possible to connect the regeneration means 11, in particular the electronic box, to the terminals of the battery 1 to be regenerated.

Acoustic emission measuring means comprising one or more acoustic emission sensors 13 are arranged on the surface of the battery 1. A sensor may be sufficient in the case of a unitary accumulator, and one or more sensors per element in the case of a storage battery can be used. Piezoelectric sensors can be used. They allow when they are brought into contact on the accumulator to detect elastic waves propagating within said accumulator. Contact between the sensor and the battery surface should be done with good acoustic coupling. The sensor can be glued, fixed or attached by maintaining a pressure between the sensor and the surface of the electrochemical system. The sensor may also be positioned on the electrical connection terminals of the system or the inter-element connections of a battery pack, the metals of these connections constituting waveguides. Contact grease allowing good acoustic coupling can also be used. The acoustic emission measuring means generally comprise an amplifier (not shown), commonly called pre-amplifier, for amplifying the signal. Such a pre-amplifier can also be integrated with the regeneration control means 14. The regeneration control means 14 comprise means for acquiring and processing the acoustic signals connected to the acoustic emission sensors 13. The means for control 14 may include a memory for storing data from acoustic measurements and their processing, and for storing a pre-established acoustic signature library. The acoustic signal acquisition and processing means comprise calculation means for processing the data. The regeneration control means 14 may also include display means for displaying the data acquired and processed, and include a user interface for acquiring the data. Advantageously, a computer is used as means for controlling the regeneration, and can be used by the regeneration means (control of the electronic box, user interface, etc.). Preferably, the regeneration device according to the invention comprises regeneration means implementing an electrical regeneration of the battery.

This configuration makes it possible to start the regeneration phase as such at the same time as the diagnostic phase. According to the invention, an acoustic emission response of the electrically biased accumulator is measured and this measurement is used to determine at least one regeneration characteristic of the accumulator. The regeneration control means 14 may thus comprise means for determining regeneration characteristics of the accumulator. The acquisition and processing means and the memory of the regeneration control means 14 may be part of the means for determining regeneration characteristics. It is thus possible to automatically determine if the battery has undergone a reversible aging phenomenon, and is not totally out of order, in order ultimately to determine whether the tested battery is regenerable or not. The acoustic emission technique may also be used according to the invention to monitor the evolution of the regeneration, and in particular to determine the end thereof. A regeneration characteristic of the accumulator can therefore in particular be the regenerable character or not of the accumulator, determined from the identification of reversible degradation phenomena by regeneration, or parameters making it possible to follow the evolution of the regeneration and in particular to determine the end.

According to the invention, the regeneration control means 14 may comprise means for analyzing an internal state of the accumulator, such as its operating state which indicates whether the accumulator is in working condition or not ( out of order or in working order). The analysis of the internal state of the accumulator can be performed during the initial diagnostic phase, or during regeneration as such. These analysis means may include the acoustic signal acquisition and processing means and the memory of the control means 14, or other means such as means for measuring and analyzing voltage, capacitance, impedance.

According to one embodiment of the invention, the regeneration may comprise, in addition to the use of acoustic measurements, another complementary electrical measurement system of the electrochemical impedance measurement type.

Measurement of acoustic emission The use of acoustic emission is known in the field of batteries, in particular to detect the failure of a battery or to estimate an internal state. The acoustic emission technique makes it possible, for example, to detect the failure of a battery, whereas this failure is not yet detectable by electrical measurement. The present invention uses the acoustic emission measurement technique to perform an initial diagnostic phase determining whether the battery can be regenerated, and to determine the end of the regeneration process as such. Such a technique applicable to batteries is for example disclosed in the French patent application FR 2,980,850, the description of which is partly reproduced below. The technique of acoustic emission Acoustic emission is defined as a phenomenon of creation of transient elastic waves, resulting from local micro-displacements internal to a material 15 (according to ASTM standard E1316-05). The term acoustic emission also refers to the discipline that uses this phenomenon. An acoustic emission source is therefore the place of an energy dissipation by sudden variation of the stress field of a material. During this variation, the energy is mainly dissipated in the form of an elastic wave, the rest being dissipated by thermal dissipation or dislocation displacement, which can in turn create elastic waves. The waves thus created move within the material to its surface. The signal measured by a sensor located on the surface of the material is much smaller than that emitted at the source. It strongly depends on the nature of the emissive phenomenon and the nature of the material, largely responsible for its attenuation. When a material is physically stressed, each change in internal structure becomes a source of acoustic emission and gives rise to transient elastic waves propagating within the material. The detection of these signals, called bursts, thus makes it possible to detect in real time the existence of these evolutions. The evaluation of the importance of these evolutions and their nature can be conducted by analyzing the acoustic parameters of the bursts emitted. Once the constraint has passed, the acoustic emission no longer makes it possible to detect after the event the evolution that the material has undergone. Many phenomena can cause acoustic emission within a material. These include: plastic deformation, dislocation movement, twinning, grain boundary slip, Piobert-Lüders band formation, fracture of inclusions or intermetallic compounds, phase transformation (martensitic), initiation and propagation of cracks (static stresses, fatigue, etc.), embrittlement by hydrogen, localized corrosion (stress corrosion, pitting, crevice), damage to composite materials and concrete (microcracking). matrix, delamination, interface breaks, fiber breaks, etc.), friction, mechanical impact. In this (non-exhaustive) list of phenomena, there are several electrochemical processes such as the dissolution of the metal or the release of gaseous dihydrogen. Thus, most electrochemical phenomena occurring simultaneously in a material constituting an electrode of any electrochemical system (such as a battery) can be detected and identified by acoustic emission. Thus, coupled with an electrochemical measurement, the acoustic emission makes it possible to obtain additional information in the measurement of the current and the potential and leads in particular to the discrimination of the different processes taking place on the electrode. Instrumentation for the acoustic emission technique The acoustic emission technique requires specific instrumentation to measure the elastic waves emitted by a material. At least one sensor 13 is placed directly in contact with the surface of the sample 1 studied. A salvo created by a micro-displacement must first propagate in the material before reaching its surface. The sensor records the mechanical waves thus created and transforms them into an electrical signal which is pre-amplified and then recorded by an acquisition system. Different obstacles or natural filters clearly change the nature and intensity of the signals as they propagate, among which the first is the material itself. The choice of the sensor must be adapted according to the type of source to be studied, the environmental characteristics, the sensitivity and the frequency filtering induced by the propagation in the material.

The sensors most commonly used in acoustic emission consist of a piezoelectric ceramic which, when in contact with the material, will convert the mechanical waves into electrical signals. The good transmission of mechanical waves between the material and the sensor is generally provided by means of an acoustic coupling such as a silicone grease. At the output of the sensor, the voltage levels generated are very low; this is why it is necessary to amplify, possibly to filter in frequency (filter pass band) and to condition the signal (adaptation of impedance for the transport of the signal, sometimes on long distance) with the help of a preamplifier. This specific instrumentation minimizes disturbances induced by the surrounding environment. The acquisition system finally allows management of the system configuration (acquisition parameters, graphical type visuals, location), but also the extraction analysis of acoustic emission characteristics when the burst is digitized. Data storage is also provided by the acquisition system.

Acoustic emission parameters The acoustic emission is done in bursts and therefore in a discontinuous manner. Each salvo has its own parameters which depend on the phenomena which are at the origin. The digital acquisition of these bursts, which constitute an acoustic emission response, makes it possible to determine these parameters and thus to classify the events recorded in families which each correspond to a disturbing phenomenon within the material. A typical acoustic emission burst is shown in FIG. 2 accompanied by the parameters calculated by the acquisition and processing system of the acoustic signals of the control means 14. Before any measurement, it is necessary to define a detection threshold ( in dB) below which no salvo will be measured by the detection system. It allows to define the beginning of a salvo during its first crossing and thus the beginning of the calculations for the characteristics of acoustic emission. It is also necessary to define time parameters in order to best achieve the measurement. The analysis parameters extracted from digitized bursts can be very numerous and are chosen by the operator. A non-exhaustive and non-exhaustive list of parameters can be defined as follows (FIG. 2): maximum amplitude (A in dB): maximum amplitude of the signal during the entire duration of the burst, - rise time (tm in ps): interval time between the first time the threshold is exceeded when the maximum amplitude is reached, duration (dm in ps): time interval between the first time out (D1) and the last exceedance (Dn) of the burst threshold , - number of strokes (without dimension, c): number of times where the amplitude of the signal exceeds the value of the threshold, energy (absolute energy or energy of EA, 1 aJ = 10-18 J): the integral of the Signal squared over the duration of the burst, peak frequency or maximum frequency (kHz): maximum frequency of the discrete Fourier transform of the burst, partial power (kHz): energy distribution per frequency band. This parameter allows the adjustment of the frequency bands to characterize the spectral energy distribution of each burst. The values are represented as a percentage of the total energy of the burst. The spectral density of energy is always calculated from a kilo point extracted from the digitized waveform. It is then possible to perform statistical processing of these bursts using signal processing software. This makes it possible to discriminate the different families of events (populations) having common parameters and thus to determine what phenomena occur within a material, as well as their duration and importance, when these are clearly identified and correlated. to a family of events. General Principle of the Regeneration Process According to the invention, the regeneration process may include a regeneration comprising at least one electrical step (ie, falling under electrical regeneration methods and mixed electrical / chemical methods) or one or more chemical steps alone. According to the invention, the regeneration method consists of sending electrical signals to the accumulator, measuring an acoustic emission response comprising a series of acoustic events resulting from said signals, and determining from this response of acoustic emission at least one regeneration characteristic of said accumulator. In this way, the regeneration can be controlled by determining whether the accumulator can be at least partially regenerated (diagnostic phase) and / or by monitoring the evolution of the regeneration by acoustic emission measurement, in particular by following the evolution of the regeneration. regeneration to determine the end. FIG. 3 schematically illustrates the principle of the method according to the invention, whatever its mode of implementation.

According to the invention, the method of regenerating an electrochemical energy accumulator, such as a battery, comprises the steps detailed below. Acoustic emission measurement (step 31) This step 31 marks the beginning of the diagnostic phase 30 of the accumulator.

It is carried out after an initial step of setting up the acoustic emission measuring means: on the one hand, the cables 12 are connected to the battery to be regenerated 1, and on the other hand the sensors 13 are placed on the surface. 1. Depending on the type of battery used, one or more sensors may be installed. Thus, a single sensor for a unitary accumulator (a single cell) may suffice, and one or more sensors per element may be used in the case of a battery consisting of a pack of cells. For example, for a 12V starter battery consisting of 6 inseparable series elements, it will be preferable to position one sensor per element. FIG. 4 illustrates such an arrangement where a lead battery comprising six electrochemical cells 31 to 36 is provided with six sensors 13, each being placed on a cell. The specific arrangement of the sensors on the different cells illustrated in Figure 4 is an example, and the sensors can be placed differently to perform the measurement. The acoustic emission measurement performed during this diagnostic phase consists in sending to the accumulator 1 electrical stresses current and / or voltage. The electrical stresses can be current profiles dedicated to the diagnosis, or be current profiles for regeneration, also called regeneration profiles in the present description. This last case is applicable if the regeneration comprises an electric step (electrical or mixed electrical / chemical regeneration method) and the diagnostic phase then corresponds to a start of regeneration as such of the accumulator. The sensors 13 then record the elastic waves produced by the accumulator under stress and transform them into electrical signals, forming an acoustic emission response. Acquisition, processing, and identification of the acoustic events measured (step 32) The diagnostic phase of the regeneration process continues with a step of acquisition, processing and identification of the measured electrical signals, also called acoustic events. in the present description. Once recorded by the sensors 13, the electrical signals which can be pre-amplified and which are recorded by the acquisition and processing system of the control means 14 are acquired. During this acquisition, the system acquisition and processing allows to filter the acoustic signals from the external environment and whose parameters are very different from those generated by the electrochemical system. These acoustic signals are then processed and analyzed, and specific acoustic signal populations are identified using the control means 14. This step consists of discriminating the different populations of acoustic events. known, that is to say, filled in a pre-established digital library, from their most characteristic and most identifiable common parameters (for example frequency, amplitude, etc.) and to account for them. For this purpose, a library of acoustic signatures previously established and stored in the memory of the regeneration control means 14 is used. This acoustic signature library groups together the acoustic signatures of various degradation phenomena that an accumulator of a type can undergo. given (for example a lead-acid battery). By acoustic signature of a degradation phenomenon is meant a parameter or a set of parameters extracted acoustic emission signals whose values are characteristic of a phenomenon attributed to a given degradation. This parameter or this set of characteristic parameters differs from all the other parameters of the other acoustic signals measured. The relation between a given degradation phenomenon and this parameter or set of parameters is previously established during the constitution of the acoustic signature library, described below. The parameters of the acoustic signals are of the medium or maximum frequency type, the duration of the signal, the rise time of the signal, the number of the signal, the signal amplitude, the signal energy, etc., but any other parameter or any combination of parameters can be used to determine the acoustic signature of a given degradation phenomenon. In practice, the treatment of acoustic events consists in highlighting the types of parameters, chosen from known parameters of the library, which can characterize each acoustic event. It is then analyzed whether these characteristic parameters of an acoustic event correspond to an acoustic signature of the library. When an acoustic event is identified as belonging to a known acoustic event population (acoustic signature recorded in the library), then it is counted. By doing so, it is then possible to display populations of acoustic events that may correspond to the acoustic signatures recorded in the acoustic signature library. By way of example, FIGS. 5A and 5B show the characteristics in energy E (in attojoule aJ). and in duration of the signal D (in microseconds) acoustic events recorded on two different elements, respectively the elements A and B, of the same lead battery subjected to an electrical stress. Each point represents an acoustic burst projected along the energy and duration axes. Element B has, compared to element A, a population of additional events distinguished from other populations by higher energy and burst duration parameters (surrounded population). This population corresponds to a phenomenon of mechanical degradation having caused an internal short-circuit in element B, element A being still functional. In this case, the acoustic emission measurement makes it possible to identify the presence of this type of irreversible internal degradation.

Note that the device and method according to the invention can use several libraries, each being for example attached to a given type of battery (lead-acid battery, lithium-ion batteries etc.). In this case, the operator can indicate, preferably during the first measurement step (for example when making the connections) the type / model of battery that he wishes to regenerate so that the appropriate library is automatically selected. According to the invention, the operator can also add libraries previously established in the memory of the regeneration control means.

The library includes acoustic signatures that can be grouped into two main categories depending on whether the degradation phenomenon that it characterizes is reversible by regeneration or is not reversible by regeneration. It is possible that the same degradation phenomenon is characterized by several acoustic signatures. Identification of the degradation phenomena making it possible to conclude on the regenerable nature of the accumulator (step 33) This step consists in detecting whether there is at least one acoustic signature characteristic of a degradation phenomenon in the various populations of acoustic events. established in the preceding step, and if necessary to identify the degradation phenomena or phenomena experienced by the accumulator so as to determine whether the accumulator can be regenerated. This step makes it possible in particular to quantify the proportion of one degradation phenomenon with respect to another by comparing the cumulative acoustic activity of each identified event population with the acoustic activity of the other populations. This step corresponds to the finalization of the diagnostic phase 30, at the end of which the diagnosis is made on the regeneration of the accumulator and whether or not the regeneration of the accumulator is carried out. To detect if there is at least one acoustic signature among the different populations of acoustic events identified for the accumulator tested, and if necessary to determine what type of degradation phenomenon has undergone the accumulator, the parameter or the group of parameters characteristic of each acoustic event population identified with each prerecorded acoustic signature of the library. The comparison can lead to the identification of one or more acoustic signatures of degradation phenomena of the accumulator. If no acoustic signature is detected, the operator is informed of a signal processing problem. The accumulator can then be considered as being non-regenerable (step 4), and be discarded, or be treated again by the system according to the invention by modifying the measurement parameters and / or data processing.

In case of identification of at least one acoustic signature, it is determined: (a) if the accumulator is not out of order, that is to say that no acoustic signature of an irreversible phenomenon having caused the out of use of the accumulator is identified (eg a short circuit); (b) if said at least one acoustic signature is that of a reversible degradation phenomenon by regeneration or an acoustic signature of the regeneration process itself in the case where the regeneration has started with the application of regeneration profiles at the Measurement step 31. If both conditions (a) and (b) are satisfied, then proceed to step 35 and proceed, or continue if the regeneration has started with the application of profiles. regeneration in the measuring step 31 (may be the case with a regeneration comprising an electrical step), the regeneration of the accumulator as such. This case (two conditions (a) and (b) satisfied) does not exclude that acoustic signatures of irreversible phenomena have been identified. However, they must not have rendered the battery out of service to allow regeneration. In this case, it is possible to estimate the part taken by each of the phenomena by analyzing the cumulative acoustic activity and inform the operator. For example, if both conditions (a) and (b) are satisfied and other acoustic populations of irreversible phenomena are identified (such as corrosion or degradation), the battery is partially regenerable in the sense that only the sulfation can be corrected. Nevertheless, the regeneration will improve the performance of the system, without allowing the accumulator to regain its initial performance since irreversible phenomena have occurred. At this stage, the operator may be free to decide whether or not he wishes to attempt a partial regeneration of his battery on the basis of information provided for example by means of the display means of the regeneration control means. 14. The information can be of the type: "reversible deterioration x%, irreversible damage y%" or directly correspond to the percentage of each of the phenomena presented in Table 1. In the opposite case (non respect of the two conditions (a) and (b): battery out of use or not out of use but without reversible phenomenon identified), the battery is considered impossible to regenerate, and the operator is informed (step 34). The battery can be discarded. To summarize, the following three situations may arise in case of identification of at least one acoustic signature: (1) out of service accumulator (HS): identification of the acoustic signature of at least one irreversible phenomenon which led to the deactivation of the accumulator; (2) non-HS accumulator and identification of at least one acoustic signature of a reversible degradation phenomenon by regeneration (with or without other acoustic signature (s) of irreversible phenomena that have not resulted in the putting out of use of the accumulator); (3) Non-HS accumulator and no identification of at least one acoustic signature of a regenerative reversible degradation phenomenon (with or without other acoustic signature (s) of irreversible phenomena that have not resulted in the putting out of use of the accumulator). Regeneration is impossible in the case of situations (1) and (3) which lead to step 34. Only situation (2) makes it possible to conclude that the accumulator is regenerable, and then goes on to step 35. In the case of a lead-acid battery, it is possible, for example, to distinguish four main types of degradation, of which only one, the sulphation is reversible by regeneration. Table 1 below outlines these main types of degradation (Robert and Alzieu, "Lead Accumulators", Engineering Techniques): Type of Causes Consequences Character reversible degradation Sulfation - Misuse / Alteration of electrical conduction YES maintenance - Accelerated by stratification of the electrolyte Corrosion of the positive electrode - spontaneous oxidation - short circuit NO at rest and in charge - alteration of the - Accelerated by the electrical conduction stratification of the electrolyte Degradation of spontaneous Phenomenon short circuits NO the active ingredient Electrolyte dehydration - recombination of the acceleration of the sulphation NO gas never effective at 100% - Loss of water during overloads Table 1 Regeneration impossible or problem of identification (step 34) This step 34 consists in informing to the operator that the regeneration is impossible or that a problem of treatment during the process does not was able to establish the diagnosis of the accumulator tested. In the event that regeneration is declared impossible, the battery may be discarded.

In the case where a processing problem has occurred, the operator can again implement the regeneration method according to the invention by modifying for example, using the user interface, the signal acquisition parameters. such as the detection threshold or the time parameters to perform the measurement. - Regeneration as such (step 35) At the end of a favorable diagnosis, that is to say when the accumulator is recognized as being regenerable following the diagnostic phase 30, the regeneration as such is initiated automatically or through an operator informed that the regeneration is possible, for example using the display means of the regeneration control means 14. The user interface of the regeneration means 11 can then allow the operator to trigger regeneration as such. Any regeneration technique may be used for the regeneration process according to the invention. Regeneration as such of the accumulator may be based on electrical techniques intended to restore at least partly the original capacity of the accumulator by particular electrical stresses applied to the accumulator, on chemical techniques generally aimed at regenerating the battery using chemical compounds administered to the battery, on mixed techniques involving electrical and chemical steps, or any other technique applicable to the present device and method. With regard to desulfation, one of the three regeneration methods known from the state of the art can be implemented. These are electrical, chemical, or mixed electrical / chemical methods. According to the invention, a distinction is made between the case where the regeneration as such comprises at least one electrical step, which corresponds to the electric desulphation methods and the mixed electrical / chemical methods, and the case where the regeneration as such comprises one or more chemical only steps, which corresponds to chemical desulphatation methods. Indeed, the two cases are distinguished at the level of the electrical profiles applied to the accumulator by the electronic unit during the measuring step 32. In the first case (at least one electrical step), since the regeneration comprises less an electrical step, the beginning of the diagnostic phase can also correspond to the beginning of the regeneration as such with the application of electrical regeneration profiles, for example pulsed current profiles of high amplitude and short duration. In this case the electrical signals sent to the accumulator, used to make an acoustic emission measurement and determine at least one characteristic of the regeneration from said measurement, at least partially regenerate the accumulator. Thus, the regeneration step 5 itself may consist of a continuation of the regeneration of the accumulator initiated at the measurement step 31. In the second case (one or more chemical only steps), the solicitation of the accumulator 1 via the electronic box of the regeneration system 11 is only by electrical profiles dedicated to the diagnostic phase. Regeneration as such only begins at step 35 according to the general scheme of FIG. 3. According to one embodiment, the regeneration of the accumulator is based on an electric regeneration technique. According to this embodiment the regeneration as such comprises at least one electrical step. Electrical desulphatation techniques for lead batteries are known from the state of the art, for example. Desulfation is caused by the application of electrical pulsations. These pulsations can be of varying intensity depending on the chosen technique. In general, they are strong, of the order of 120-300 A, and more rarely weak, of the order of 8 A. Prior to the application of electric pulsations, distilled water and / or 20 l The electrolyte can be added to the accumulator 1 in order to adjust the density of the electrolyte. Such desulfation techniques apply if the battery has not been mechanically damaged, and preferably if the battery has a minimum capacity. Examples of such electrical desulphatation techniques are for example described in patent application WO 12026071 or US 2010/117606. According to another embodiment, the regeneration of the accumulator is based on a technique of mixed electrical and chemical regeneration. As for regeneration by an electrical technique, the regeneration according to this embodiment comprises at least one electrical step. The regeneration then differs from the regeneration by an electrical technique mainly in that a chemical compound is added to the accumulator once the accumulator recognized as being regenerable. This chemical additive allows, in combination with the particular electrical stresses of the accumulator 1, to regenerate the accumulator. This chemical additive is for example added to the accumulator. During the step of adding the chemical compound, the battery is preferably not electrically biased. Regenerative electrical profiles are applied once the chemical compound has been added to the accumulator. These mixed desulfation techniques are also known from the state of the art. A chemical additive is first injected into the accumulator, for example using a syringe once the accumulator caps open. In a second step, the accumulator is connected to a desulfation machine. In the present invention it is the regeneration system 11 comprising an electronic box, which is connected to a current source, and which makes it possible to apply electrical stresses to the accumulator 1. Prior to the application of electrical pulsations and following the opening of the plugs of the accumulator during the injection of the chemical compound, distilled water and / or electrolyte may be added to the accumulator 1 in order to adjust the density of the the electrolyte. Preferably, such mixed desulfation techniques apply if the accumulator has not suffered mechanical damage, and preferably if the accumulator has a minimum capacity (existence of a threshold below which it is difficult to obtain an effective desulfation) and / or a minimum age. The chemical additive may be based on ethylene diamine tetraacetic acid (EDTA), water, organic polymer, hydrogen peroxide and / or other components. In general the chemical additive is added according to a specific volume calculated according to the capacity and the condition of the accumulator. The improvement of the capacities is sometimes optimal after several weeks of use of the thus desulfated accumulator (several cycles of charge and discharge) According to another embodiment, the regeneration of the accumulator is based on a regeneration technique. chemical, that is to say comprising one or more chemical steps without application of an electrical regeneration profile, in which case the operator who has been informed, for example by means of the means of display means regeneration control 14, that the battery is regenerable, adds a chemical additive, for example by means of a syringe, in the accumulator 1 whose caps have been opened.The quantity of chemical additive poured can be calculated in depending on the 5 parameters of the accumulator (mainly its capacity) Distilled water can also be added to adjust the level of the electrolyte.In general, at the end of the desulfate At this time, the battery is charged at 100 `Vo. This activates the additive and then reuses the battery. The chemical additive may be based on ethylene diamine tetraacetic acid (EDTA), water, organic polymer, hydrogen peroxide and / or other components. Preferably, such chemical desulfation techniques apply if the battery has not suffered mechanical or visible damage. It is preferable that the accumulator has a minimum voltage with respect to the manufacturer's voltage (existence of a threshold may exist below which it is difficult to obtain efficient desulphating). The density of the electrolyte is preferably adjusted to correspond to a given range, for example the range of 1.15 and 1.30. It may also be advantageous to control the density difference between the different cells of an accumulator, so that it is less than a threshold value, for example 0.04. Examples of such chemical desulfation techniques are, for example, described in US patent application 2009/0140695. During regeneration as such, and regardless of the regeneration technique employed, the acoustic activity (number of events) or cumulative energy corresponding to at least one mechanism related to this regeneration is recorded during the regeneration process. time, by means of acquisition and processing of the regeneration control means 14. It is then determined that the regeneration is complete when a given criterion is reached. For example, when the acoustic activity corresponding to the regeneration disappears or goes below a certain threshold. Another criterion may be the appearance of a population of acoustic events corresponding to a regenerated battery. Finally, when the regeneration itself causes identifiable degradations by a population of acoustic events, a stopping criterion can be defined by the appearance of this population or the passage of the acoustic activity related to this population above a certain threshold. Indeed, in this case, the treatment itself can reduce the performance of the battery if the degradation processes are higher in quality regeneration processes. When this criterion is reached, the regeneration is stopped and the operator is warned that regeneration is complete. The end of the regeneration decision can be taken automatically by the control means 14 controlling the cessation of the regeneration by the regeneration means 11 or by the operator who acts directly on the regeneration means 11. This monitoring is carried out as soon as possible. beginning of the regeneration as such of the accumulator, which may or may not coincide with the beginning of the diagnostic phase 30, for example in the case where a method of electrical or mixed electrical / chemical regeneration is employed. In the case where the regeneration as such is chemical, the monitoring of the regeneration by measuring an acoustic emission response of the accumulator is carried out without applying an electrical profile to the accumulator. At the end of regeneration as such, the regenerated accumulator can be subjected to particular treatments, for example to be cooled, stabilized, change of defective elements, etc., and to be the subject of additional conventional tests for 20 check the new state of the accumulator (discharge test to check the capacity of the desulfated accumulator, tests of voltage, current, electrolyte level etc.) Method of establishing a library Acoustic signatures The establishment of an acoustic signature library is enabled by a clear identification of different phenomena of degradation of the accumulator and their correlation to families of acoustic events. These event families are defined based on one or more common parameters that have characteristic values. We will speak of acoustic signature of a degradation phenomenon given once established the relationship between a family of acoustic events and said degradation phenomenon. The library or libraries used according to the invention are preferably established before any implementation of the regeneration process.

A library may be typical of a given type of accumulator or group acoustic signatures of different types of accumulator. Different ways to establish such a library of acoustic signatures can be implemented. For example, it is possible to proceed as described below with reference to FIG. 6. Firstly, a laboratory test campaign is carried out on a batch of accumulators 6 in order to measure the acoustic signals produced by the batteries. accumulators when they are solicited (in operation, charging, discharging, pulse current, etc.) and this according to different controlled degradation states, temperature, state of charge, etc. The accumulators 6 of the batch are preferably of the same type as the accumulator that it is desired to regenerate. Acoustic emission signals (S) corresponding to the n states of degradation (E) are obtained. The different degradation states of the accumulators can be obtained by accelerated aging campaigns in which one degradation phenomenon is favored over another. For example, sulfation is known to be preferred for moderate temperatures and for low sulfuric acid concentrations or low states of charge, whereas corrosion is preferred for high load states and high temperatures. Alternatively, used accumulators can be used. In a second step 61, a statistical processing of the recorded signals is carried out starting on the one hand by grouping them by population of signals (pop S), each population having common parameters. The parameters of the acoustic signals are of the type: average or maximum frequency, duration of the signal, time of rise of the signal, number of blow of the signal, amplitude of the signal, energy of the signal, etc. Signal populations for n states are obtained. When the populations of acoustic events have simple distributions, it is possible to discriminate them by representing these populations along two-dimensional axes, the two dimensions being acoustic parameters (for example the frequency of acoustic bursts according to their amplitude). . This simple processing 30 can make it possible to display sets of events grouped into populations of similar parameters. In cases where populations have complex distributions, a principal component analysis can be performed. Each acoustic emission signal is represented in a multidimensional space by a vector whose coordinates are the different acoustic parameters. For example, an acoustic signal can be defined by a vector of type: S = (f (amplitude, energy, number of strokes, duration, rise time, etc.)) The separation of all these vectors into several Event populations can then be performed as follows: deletion of parameters giving no information on the physical nature of acoustic emission signals, such as the threshold and the acquisition channel, removal of the correlated parameters when the information of only one of several parameters is sufficient. projection in normalized eigenvectors: the standard linear projection on a (2D) plane of uncorrelated parameters renders the system dimensionless and makes it possible to maximize the dispersion of the vectors. The dimension of the data network corresponds to the number of projected parameters. Thus, 2 vectors with similar characteristics are represented in the same region of space. For example, Figure 7A shows a projection of acoustic data along the normalized eigenvectors VO and V1. The events are divided into two distinct directions corresponding to two populations of pop events () (black dots) and papi (gray dots).

The separation of the set of acoustic emission signals into several classes can be performed from a statistical processing. There are different methods of classification such as the use of a simple iterative algorithm (K-means), which tends to minimize the squared difference between each point for a given number of populations. When an event population is clearly determined by the software used, it can be deleted from the file and the statistical processing taken over the remaining part. In a third step 62, a population of pop events S is assigned to a state of degradation E of the accumulator. For example, a population of events can occur when an irreversible phenomenon of corrosion degradation can lead to a failure of the accumulator, and thus characterize the irreversible phenomenon of degradation by corrosion. The various degradation phenomena listed in Table 1 may, for example, be linked to specific populations of acoustic signals, these being then characteristic of the various degradation phenomena of the accumulator, some of which may be reversible by regeneration of the accumulator. , others are irreversible, even causing the battery to be shut down (eg short circuit). This step implies that the state of degradation of the accumulator is known, so that the link between the population of acoustic events and the given state of degradation is established. This involves an analysis of the clinical signs of the accumulator, which can for example be observable directly or indirectly. It is thus possible to use measurement techniques such as microscopy (optical or scanning electron) or X-ray diffraction or measurement of the adhesion of the active material to the collector grates. For example, in the case of a mechanical degradation of the electrode causing the decohesion of the active ingredient and may lead to a short circuit, the disassembly of the battery and the measurement of the mechanical strength of the active ingredient on the gates by a simple adhesion test (using for example a grid comb) can make it possible to identify this phenomenon of degradation compared to the mechanical strength of a new electrode. In a fourth step 63, it is sought to identify the parameters of the signals or the combinations of parameters of the signals that are different from the parameters of the other signals among the populations that correspond to the various degradation states of interest. For example in Figure 5B, the event population related to the mechanical degradation of the active ingredient is characterized by burst times associated with energies both higher than other populations. Thus, with steps 62 and 63, an acoustic signature characteristic of a given degradation state is determined by calibrating a relationship between at least one of the parameters common to at least one population of acoustic signals and a state of degradation. given the accumulator of the tested batch, by means of the analysis of the values of one or more common parameters of the acoustic signal populations obtained for each of said degradation states. The set of characteristic acoustic signatures determined then constitutes a library that can be stored in the memory of the regeneration control means 14 of the device according to the invention.

According to the invention, it is also possible to perform the processing of acoustic events without a statistical processing stage of the acoustic events using signal processing software. This is the case, for example, when a single parameter differs from the other parameters in the recorded acoustic events, this parameter alone making it possible to characterize a given type of degradation. The acoustic signature is in this case very simple. For example, in the case where the acoustic signature of a given degradation phenomenon corresponds to a population with a high burst energy (and greater than the energy of the other bursts), a device capable of detecting the exceeding of a threshold of d Reference burst energy can detect the corresponding type of degradation, with the high energy being a sufficient acoustic signature of the state. Example A used and charged battery, whose capacity was initially measured at 6.5 Ah, is subjected to regeneration by a device using an electrical method alone for about 27 hours. An acoustic emission sensor has been placed on one of the elements of the battery and the acoustic events are recorded during the entire regeneration step. After regeneration, the measurement of the battery capacity indicates that the battery has doubled to 13 Ah.

All the acoustic events recorded during this test are represented on the graph of FIG. 7B according to their duration D in microseconds (in abscissa) and their energy in attojoule (in ordinate). The vectorization of these events and their projection in normalized eigenvectors VO and V1 (FIG. 7A) according to the method described above makes it possible to identify two populations of pop () (black) and popi (gray) acoustic events. The plot of the cumulative acoustic activity of these two populations over time during the regeneration test shows that the population of pop () acoustic events has a very high activity during the first 25,000 seconds and then reaches a plateau. On the other hand, the popi population has a moderate activity during the first 30,000 seconds and then increases exponentially until the process stops at about 100,000 seconds. The analysis of this test shows that the pop population () corresponds to a population of events directly related to the regeneration of the battery, which had no significant effect after 25,000 seconds. The pop-I population has not been formally attributed to a specific phenomenon in the context of this test but could correspond either to a phenomenon related to the battery overload (the current pulses charging the battery) or to a mechanism degradation linked to an excessive regeneration process. It can be concluded from this example that tracking pop () event population over time is a way to control the regeneration step and determine the end of it.

Claims (14)

REVENDICATIONS1. Regeneration device (10) of an electrochemical energy accumulator (1), comprising: regeneration means (11) comprising an electronic electrical signal generating unit connected to said accumulator (1); acoustic emission measuring means comprising at least one acoustic emission sensor (13) placed on said accumulator (1); regeneration control means (14) comprising means for acquiring and processing acoustic signals emitted during the generation of the electrical signals. 2. Device according to claim 1, wherein said control means (14) comprise a storage memory of a library of characteristic acoustic signatures. 3. Device according to one of claims 1 and 2, wherein said control means (14) comprise means for determining regeneration characteristics of said accumulator (1). 4. Device according to one of the preceding claims wherein said control means (14) comprise means for analyzing an internal state of the accumulator (1), such as its operating state. 5. Device according to one of the preceding claims, wherein the accumulator is lead. 6. Device according to one of the preceding claims, wherein the control means (14) comprise an electrochemical impedance measuring system. 7. A method for regenerating an electrochemical energy accumulator (1), in which: - electrical signals are sent into said accumulator (1), an acoustic emission response comprising a series of acoustic events resulting from said acoustic emission is measured; signals; from said acoustic emission response at least one regeneration characteristic of said accumulator is determined. 8. A method of regenerating an electrochemical energy accumulator (1) according to claim 7, wherein: - it is determined whether the accumulator (1) can be at least partially regenerated. 9. Regeneration process of an electrochemical energy accumulator (1) according to one of claims 7 and 8, wherein: it controls the evolution of said regeneration of the accumulator (1) by measurement of acoustic emission . 10. A method of regenerating an electrochemical energy accumulator (1) according to one of claims 7 to 9, wherein: a pre-established characteristic acoustic signature library is used to determine at least one regeneration characteristic of said accumulator. . 11. Regeneration process of an electrochemical energy accumulator (1) according to one of claims 7 to 10, wherein: a treatment of acoustic events by grouping them by signal populations having common characteristic parameters. The method of regenerating an electrochemical energy accumulator (1) according to claim 11, wherein: comparing said signal populations with characteristic acoustic signatures grouped in a preset acoustic signature library; different populations of acoustic events characteristic of different degradation phenomena experienced by the accumulator at the end of the comparison step are identified; anddetermining the respective share of each of said phenomena by comparing the cumulative acoustic activity of each of the different populations of identified events. 13. A method of regenerating an electrochemical energy accumulator (1) according to one of claims 7 to 12, wherein the electrical signals sent to said accumulator (1) at least partially regenerate said accumulator (1). 14. A method of regenerating an electrochemical energy accumulator (1) according to claim 10, wherein the library of characteristic acoustic signatures is established prior to sending electrical signals to the accumulator (1) for the measurement of an acoustic emission response, according to the following steps: a batch of electrochemical energy accumulators (6) of the same type as the electrochemical energy accumulator (1) to be regenerated is selected; for different degradation states of at least one accumulator of the batch, electrical signals are sent to said accumulator of the batch, at least one acoustic emission response comprising a series of acoustic signals resulting from said electrical signals is measured, and said signals are recorded hearing aids; the acoustic signals recorded are processed by grouping them by signal populations having common parameters; and determining at least one acoustic signature characteristic of a given degradation state by calibrating a relationship between at least one of the parameters common to at least one population of acoustic signals and a given degradation state of said accumulator of the batch, by means of an analysis of the values of one or more common parameters of the acoustic signal populations obtained for each of said degradation states.