JP7728965B2 - Estimating the health status of electrochemical devices - Google Patents

Description

This disclosure relates to the field of energy storage in electrochemical devices, particularly battery-type devices, and the second life of such devices.

Such electrochemical energy storage devices can be used in any electrical device, such as a smartphone, laptop, or in any electrical system, such as an electric vehicle (EV) or a battery energy storage system (BES), especially for grid operators.

More generally, electrochemical devices of the types described above may comprise primary or rechargeable (secondary) batteries.

The rechargeable battery may be a lithium-ion battery, a nickel-metal hydride battery, a nickel-cadmium battery, a lead-acid battery, a solid-state lithium-ion battery, a solid-state lithium metal battery, a sodium-ion battery, a solid-state sodium-ion battery, a solid-state sodium metal battery, or a sodium-sulfur battery.

The primary battery may be an alkaline battery, a lithium battery, a lithium-air battery, or a zinc-air battery.

The present invention can therefore be applied to any device or system in which an electrochemical energy storage device of the type described above is present.

For such storage devices, there is a desire to obtain the state of health (or SOH) of the electrochemical energy storage device. Known techniques may utilize measuring the charge/discharge capacity of the device, measuring the impedance of the device, or looking at historical data for the device.

SOH estimation can vary according to the nature of the electrochemical device (rechargeable or primary). In the case of rechargeable batteries, most of the electrochemical reactions are usually reversible, but some irreversible reactions (side reactions) may occur. The recharge capacity of the battery decreases due to these irreversible reactions. However, the materials inside the battery cannot be analyzed without disassembling and destroying the device. There are specialized techniques for analyzing the materials inside the battery, for example, using high-energy X-rays, but these techniques are very expensive and are not used in practice.

Almost always, a battery's state of health (SOH) is estimated based on the charge/discharge capacity and the variation trend in its operating voltage. In these cases, historical data about the battery is needed to estimate its SOH. When wanting to know a battery's SOH without history, the preferred method is to charge/discharge the battery in question several times to estimate its SOH. This type of estimation is necessary, for example, when an electric vehicle (EV) battery is reused as a second-life battery in a BES type system. Normally, charging and discharging takes a long time. For large battery systems such as EVs and BESs, a large amount of electrical energy is required and the process takes a long time.

Irreversible reactions are primarily due to electrolyte consumption. Focusing on this characteristic, changes in electrolyte composition have been proposed as an SOH indicator. However, the proposed measurement requires cooling to -40°C relative to ambient temperature. In addition, precise linear temperature control is required to capture the electrolyte's melting point and/or glass transition temperature. The batteries tested are typically limited to small cells with dimensions of 40 mm x 20 mm x 3.5 mm. If the proposed procedure were applied to larger batteries used in EVs and BES, a bulky and expensive measurement system would be required. Cells are typically assembled into modules. Therefore, users would have to disassemble the modules to remove the cells before applying these measurements. Such a procedure is impractical. Furthermore, the proposed procedure can only be applied to batteries containing electrolytes with melting and/or glass transition temperatures. Recently, solid electrolytes without melting and/or glass transition temperatures have been proposed for new-generation batteries. In this case, such SOH estimation is not feasible.

In the case of lead-acid batteries, measuring the battery's impedance is commonly applied to estimate its SOH, since lead-acid batteries are generally used at the same state of charge (SOC). However, other rechargeable batteries are used at different SOCs. Generally, the change in battery impedance depends on the SOC. In this case, therefore, it is necessary to charge or discharge all batteries to the same SOC in order to compare impedance values and estimate their SOH. In all cases, charging or discharging the batteries is required.

In the case of primary batteries, some of these batteries exhibit a voltage drop proportional to their SOH. On the other hand, batteries such as zinc-air batteries exhibit a flat voltage profile during most of the discharge process. In this case, it is difficult to estimate the SOH reliably based on the voltage. Therefore, it is not possible to estimate the SOH of a primary battery from the fluctuations in battery voltage.

This disclosure improves this situation.

1. A method for estimating the state of health (or "SoH") of an electrochemical device, comprising:
- recording the thermal response of the electrochemical device to a heat supply applied to the electrochemical device;
- measuring, in the thermal response, at least one parameter representative of the thermal inertia of the electrochemical device;
- inferring an estimate of the state of health of the electrochemical device from measurements of said parameters.

It has been observed that the thermal response of electrochemical devices changes according to the age of the device, and in particular its state of health. The exemplary embodiments and results presented below clearly show the effect of state of health on the device, in particular its thermal inertia. In the examples presented below, the more deteriorated the device, the greater its thermal inertia. This observation can be explained by the fact that the irreversible reactions that the device undergoes over its lifespan (deterioration of its state of health) change the materials present in the device, and these new phases then have different thermal inertias.

The results presented below show a correlation between a parameter characterizing the thermal inertia of a device and its state of health, so that by measuring this parameter it is possible to infer the state of health of the device or even predict its remaining life under similar conditions of use.

To do so, it is sufficient to supply heat to the device and measure this parameter that characterizes its thermal inertia. The amount of heat supplied can be "positive" (heating the device) or "negative" (cooling), and the main idea is to measure the thermal response of the device to this heat supply and to infer from this the measurement of the above-mentioned parameter. The heat supply can be generated by a source (cold or hot) external to the device. Alternatively, the heat supply can simply result from the normal operation of the device, for example during charging phases, since the device naturally heats up during these phases.

The thermal response of the device can be obtained after (and/or starting from) the application of a heat supply, for example, by recording its temperature fluctuations over time. Alternatively, the thermal response can be obtained by measuring the heat flux using a device comprising a Peltier module.

It is thus proposed to measure the thermal response of the device (possibly via a heat sink). This response can be expressed by its temperature or by the heat flux between the device and the sink or simply between the device and the ambient air. This thermal response can be after the application of an external heat source (hot or cold) or after a specific operating phase of the device (e.g. charging). The state of health of the device is inferred from the analysis of this thermal response.

As noted above, in the exemplary embodiment presented below, the thermal inertia of the electrochemical device being tested increases as the device's health decreases. However, for some different electrochemical devices, the trend may be the opposite.

As illustrated in the bottom graph of FIG. 2, discussed below, the measured parameters mentioned above can include the delay (denoted as t _delay ) until the thermal response of the electrochemical device begins relative to the start time of application of the heat supply.

Additionally or alternatively, after a start time of application of the heat supply, the heat supply is applied continuously and the thermal response of the electrochemical device is recorded as a function of time, and the above-mentioned measured parameters include the slope of the variation of the thermal response as a function of time (dP/dt).

Additionally or alternatively, the heat supply is applied continuously for a selected duration after the start time of application of the heat supply, and then terminated after the selected duration, and the measured parameters include the maximum amplitude (Pmax) of the thermal response of the electrochemical device.

Additionally or alternatively, the measured parameters may include the length of time (τ) for the thermal response of the electrochemical device to return to a predefined threshold (P ₀ ) after terminating the application of the heat supply, as illustrated in FIG.

In one embodiment, the electrochemical device is in operation during the application of the heat supply. While such an embodiment is not necessary for the application of the present method, it is advantageous because it is not necessary to remove the electrochemical device in order to perform measurements of the above-mentioned parameters.

In one embodiment, the measurements of the above parameters are stored in memory to monitor changes in the health of the electrochemical device over time. For example, a prediction of remaining life can be assessed in this manner.

For example, the method may enable generating a warning signal when the health state of an electrochemical device falls below a threshold. Such an embodiment may then enable, for example, predicting the end of life of an electrochemical device.

The present invention provides
A device for estimating the state of health of an electrochemical device, comprising:
at least one sensor for detecting the thermal response of the electrochemical device to a heat supply applied to the electrochemical device;
a processing circuit for measuring, in thermal response, at least one parameter representative of the thermal inertia of the electrochemical device and for inferring from the measurement of said parameter an estimate of the state of health of the electrochemical device.

The sensor described above may be, for example, a Peltier module (e.g., using the Seebeck effect) as described below by way of example, or any other means of measuring the thermal response of an electrochemical device.

In one embodiment, the device may further comprise a heat transfer device attached to the electrochemical device and configured to apply the heat supply described above to the electrochemical device. Additionally or alternatively, the heat supply may come from the operation of the electrochemical device itself.

For example, the device may further include a pad made of an electrically insulating material, and the transfer device may be configured to be attached to the first main surface of the electrochemical device by means of the pad. The transfer device may also be covered with such an insulating material and in the form of a mat that heats the electrochemical device.

The device may further include a heat sink attached to the electrochemical device. For example, the heat sink may be attached to a second major surface of the electrochemical device opposite the first major surface described above.

A heat sink can be used, for example, to measure the thermal response. It is not necessary (but is useful for small electrochemical devices to concentrate the heat flux as it is being measured).

In one embodiment, the processing circuitry described above may include memory for storing measurements of at least some of the parameters and for monitoring changes in the health of the electrochemical device over time.

The present invention also relates to a computer program comprising instructions that, when executed by a processing circuit, implement the steps of the above-described method.

According to another aspect, a non-transitory computer-readable storage medium is provided on which such a program is stored.

Other features, details, and advantages will become apparent upon reading the following detailed description and examining the accompanying drawings.

FIG. 1 illustrates a schematic example of a device for heating and measuring the thermal response of an electrochemical device, according to one embodiment. FIG. 1 illustrates a typical thermal response of an electrochemical device after heat is supplied by a heat source. FIG. 3 illustrates the thermal response (with heat flux peaks) for cells at 99.5%, 91.5%, and 87.4% SOH after heat is supplied by a heat source with a stepwise increase in power (rectangular peaks on the left side of FIG. 3 ). 10A-10C illustrate the thermal response of another electrochemical device at different states of health of the device. FIG. 10 shows the correlation between the slope of the time variation of the thermal response (dP/dt) as a function of time versus the health state of the device. FIG. 11 shows the correlation of the length of time (τ) required for the thermal response to return to normal versus the health of the device. FIG. 10 shows the correlation of the peak height (Pmax) of the thermal response to the health state of the device. FIG. 10 shows the correlation of the delay in thermal response (t _delay ) after application of a heat supply to the health state of the device. 1 illustrates a device comprising a processing circuit CT according to an embodiment of the present invention. 1 illustrates steps of a method according to one embodiment of the present invention.

In the exemplary embodiments presented below, electrochemical devices, when involving rechargeable devices, typically undergo one or more irreversible reactions followed by a loss of capacity and the onset of device degradation. Indeed, as the SOH of the device changes, an irreversible reaction occurs inside the device. This irreversible reaction creates a new phase in the device's material with different thermal properties, such as heat capacity and/or heat transfer. For example, lithium-containing deposits can be observed at the anode, particularly in the case of lithium batteries. Additionally, this irreversible reaction causes a change in the device's volume. These changes in properties can occur simultaneously or independently.

The SOH estimation for electrochemical energy storage devices presented below uses the difference in the thermal properties of the device after capacity degradation described above.

Reference is now made to Figure 1. In Figure 1:
the reference "a" denotes an electrochemical device, for example of the battery type,
- the reference "b" denotes a heating element,
- the reference "c" denotes a temperature or heat flux sensor,
- The reference "d" denotes a heat sink.

Thermal energy is applied to an electrochemical energy storage device and a thermal response is obtained to estimate the SOH of the device. Here, the energy is applied by, for example, a heating element located on the device, as shown in FIG. 1. However, this technique is not limited to specifically "heating" elements. A cooling device can also be used to change the thermal state of the device in a controlled manner. For example, the cooling device can include a Peltier effect module.

The heat or cold source can be located outside or inside the electrochemical device.

"Delivering" thermal energy (hence positive or negative) consists, for example, of applying a known value of energy to change the thermal state of the device as shown in Figure 2.

One way to do this may consist of applying thermal energy corresponding to the largest increase in thermal response. Generally, when thermal energy is applied to a device, the device's temperature begins to increase or decrease. Then, after a long period of energy application to the device, its temperature reaches an equilibrium state, where any additional applied energy is released to the environment by heat transfer between the device and the ambient atmosphere. Thus, as illustrated in Figure 2, the temperature peaks between the start of energy application and the device's equilibrium state. This point is of primary importance for comparing differences in the thermal properties of devices.

However, the procedure proposed above is just one non-limiting example. For example, to assess the change in the thermal behavior of an electrochemical device, it is also possible to compare the onset of the change in the temperature of the device (denoted as t _delay in FIG. 2).

Another method may be to compare the rate of relaxation after the application of thermal energy. Generally, the rate of thermal relaxation of a material is described by "Newton's law of cooling," where thermal relaxation has an exponential relationship with time, and the exponential term includes a heat transfer coefficient τ, which, in such electrochemical devices, appears to be proportional to the device's heat capacity C and inversely proportional to the heat transfer coefficient α. Changes in the device's volume also cause changes in heat capacity and/or heat transfer. Therefore, the SOH can be estimated by measuring the device's relaxation rate after the application of thermal energy.

Another parameter that characterizes the thermal inertia of an electrochemical device is the slope of the time variation in the thermal response (e.g., the rising slope in Figure 2).

Detection Procedures There are several options for detecting the applied thermal energy.

Various temperature sensors, such as thermocouples, thermistors, resistance temperature detectors (RTDs), or infrared thermometers, can be used where a detectable temperature change in the device can be measured after the application of thermal energy. Peltier effect modules can also be used as heat flux sensors. Such sensors measure heat flux by delivering an electrical signal (e.g., in mV) via the Seebeck effect, thus enabling the thermal response of an electrochemical device to be measured (measurements given in mV in the experimental procedures presented below as examples).

If the heat flux generated by the device is too small to alter its measured temperature, such a heat flux sensor can be used between the device and the external atmosphere or between the electrochemical device and the heat sink, indicated as "d" in the figure. Such an embodiment makes it possible to improve the signal-to-noise ratio of the measurement.

Experimental Procedure: A lithium-ion battery is used below as a sample for the test, the results of which are illustrated in Figure 3. The nominal capacity of the battery is 4000 mAh. Its size is 50 mm (width) x 90 mm (length) x 5 mm (thickness). The battery was charged and discharged at 200 mA between 4.4 V and 3.0 V. The starting point for reversible capacity was defined for 100% SOH. The battery was then continuously charged and discharged. After the first few cycles, the capacity conservation was verified under the same conditions as defined at the beginning of the experimental procedure.

After verifying the capacity, the thermal properties of the battery were measured as follows: The battery to be tested was placed in a temperature-controlled chamber set at 25°C. The battery was then kept separated (or covered) from the battery by a rubber pad (the rubber is thermally conductive but electrically insulating) and heated using a heating element, such as a heating mat, attached to the battery. The applied heat was 0.74 W for 180 seconds. The thermal response of the battery was measured using a Peltier module attached to the opposite side of the heating element. The Peltier module was then inserted between the battery and an aluminum plate (which acts as a heat sink). The difference between the maximum value of the thermal response and the base value P0 was defined as Pmax. The cooling rate was then defined using the following equation:
where P is the thermal response, P0 is the base value of the thermal response, A is a function of the applied power, t is the elapsed time, t0 is the starting point of heating, and τ is the cell cooling rate. Pmax is obtained by observing the maximum peak value P-P0.

In addition, the total volume of the battery was measured using the Archimedes method.

In the initial condition, when the SOH of the battery was 99.5%, the Pmax value was 23.0 mV, τ was 149.5 seconds, and the volume of the battery was 19.59 cm ³ .

Next, under the first experimental condition, when the battery's SOH was 91.5%, the Pmax value was 22.0 mV and τ was 160.7 seconds.

Then, in the second experimental condition, when the SOH of the battery was 87.4%, Pmax was 21.6 mV, τ was 165.8 seconds, and the battery volume was 21.07 cm ³ .

These values are referenced in the table below.

These results can be seen in Figure 3.

As shown above, as the SOH decreases, the value of Pmax decreases and the rate τ increases. The battery volume increases as the SOH decreases. Therefore, the thermal response using the proposed procedure can be used as an indicator for estimating the battery's SOH. The battery volume is also an indicator for estimating the battery's SOH and can support SOH estimation via thermal measurements.

Therefore, the proposed procedure for estimating the SOH of a device by using its change in thermal response relies on the heat capacity and/or heat transfer of its inner material. This procedure using thermal response can be applied to other devices using other detection methods.

For example, if a material in a device exhibits a corrosive reaction after use, similar techniques can be applied to determine the extent of corrosion, since the thermal response is affected by the formation of corrosive elements.

For example, if a material in a device absorbs moisture after use, similar techniques can be applied to determine the extent of water absorption, as the thermal response will be affected by the formation of elements associated with water absorption.

Therefore, such a method can be proposed to estimate the SOH of an electrochemical device during the actual operation of a system of which the electrochemical device is a part, such as an electric vehicle or a BES system. In these systems, a large number of devices are used. Generally, the SOH of the entire system can be estimated during the system operation period, such as during charging and discharging. However, when estimating the SOH of each device, disassembly of the system is required to measure the SOH of each device. Here, the proposed method can make it possible to estimate the SOH of a device without any disassembly operation, since each device can easily be equipped with a thermal source and a thermal sensor.

Such embodiments can be beneficial to operators of electrochemical storage systems by enabling them to identify and replace faulty devices in the storage system instead of replacing the entire system.

For example, after being used in an EV, an electrochemical storage device can be used as a second-life device in a BES system. This type of second-life use is encouraged to reduce the carbon footprint of the lifecycle device, increase the resale value of the EV, and reduce the installation costs of the BES.

During this secondary use period, it is necessary to estimate the SOH for each device. In some cases, there is no reliable historical data for the primary use of an EV. In such cases, it is useful to estimate the SOH individually using the implementation described above.

Furthermore, this procedure does not require any charging or discharging of the device to estimate its SOH, which is also a significant advantage due to the savings in operating time and energy. Such applications could be beneficial to car dealerships, resellers/users of second-life batteries, and energy companies.

FIG. 4 illustrates the thermal response of another electrochemical device for different states of health of the device.
- the slope of the temporal variation in the thermal response as a function of time (dP/dt) (Fig. 5)
- the length of time (τ) required for the thermal response to return to the threshold _P0 (Figure 6)
- Peak height of the thermal response (Pmax) (Figure 7)
- the delay of the thermal response after application of the heat supply (t _delay ) (FIG. 8)
Thermal inertia parameters such as these are strongly correlated with the health state of the device.

At the very least, peak height (Pmax) has been found to be a parameter with a significant correlation. In addition, its measurement error is small.

9 illustrates a processing circuit CT connected to a sensor c for measuring a thermal response to a heat supply AC that can be controlled by the processing circuit CT (references a, b, and c of FIG. 1 represent the same elements in FIG. 9). The processing circuit CT typically comprises a processor PROC cooperating with a storage memory MEM for reading therefrom and is capable of executing the instructions of a computer program of the type described above (as well as thermal response measurements, and thermal inertia parameter values at different time stamps, estimated state of health values, etc.). The processor PROC can also control the heat supply as well as the collection MES of measurements by the sensor c. Typically, with reference to FIG. 10, the processing circuit CT:
- triggering the application of a heat supply to the electrochemical device (step S1 of FIG. 10),
- controlling the sensor to record the thermal response of the electrochemical device to the heat supply (step S2);
- evaluating (S3) in the thermal response at least one parameter representative of the thermal inertia of the electrochemical device;
- From the value of this parameter, an estimate of the state of health SOH of the electrochemical device is deduced (S4).
It can be configured in this way for

The thermal response measurement
- thermocouples (to measure temperature),
- resistance temperature detectors (for measuring temperature),
- thermoelectric modules (measuring the heat flux), for example of the Peltier type using the Seebeck effect (the measurements in the above table were obtained by such means),
- thermistors (to measure temperature),
- Thermometers (e.g. infrared)
This can be done by the sensor c through various means, such as

Heat supply is, for example,
- Resistors, heating mats or pads (Kapton® type or other types)
- a circuit containing a (temperature-controlled) heat transfer fluid,
- chemical reactions,
- visible or infrared range or other laser light,
- the electrochemical device itself (e.g. when charging),
- This can be done by applying a heat or cold source, which may be the real environment of the electrochemical device. For example, in this application, when an electrochemical device is handled to move it from a warehouse where it is stored at a given temperature (e.g., 25°C) to another environment (with a different temperature), its thermal response to this temperature variation can be measured, from which the state of health of the electrochemical device can be deduced. Thus, in this case, the mere movement of the electrochemical device into an environment with a different temperature can be suitable for measuring its state of health. During this mere movement, it is therefore possible to separate, for example, electrochemical devices that can be reused in a second life from those that should definitely be recycled.

The general implementation described above allows the automaker to provide the EV user with a more accurate estimate of the SOH in this case. In addition, the automaker can provide the second user of the battery with an accurate resale value.

In the case of energy companies, this implementation allows the exact number of years of operation of the BES to be estimated. In residential applications of the BES, energy companies can provide residential users with a reliable estimate of SOH and inform them of maintenance or replacement periods for the BES.

Several tests were performed on cells of different types (different sizes, formats, and chemistries). The results show that this implementation is effective for estimating battery SOH and can be readily deployed for industrial purposes.

a Electrochemical device b Heating element c Temperature sensor, heat flux sensor d Heat sink

Claims (14)

1. A method for estimating state of health (SoH) of an electrochemical device, comprising:
recording a thermal response of the electrochemical device to a heat supply applied to the electrochemical device;
measuring at least one parameter indicative of a thermal inertia of the electrochemical device in the thermal response, the thermal inertia being defined as the time it takes for the electrochemical device to change temperature in response to the supply of heat, the thermal inertia of the electrochemical device increasing as the State of Health (SoH) of the electrochemical device decreases, and the parameter is selected from a delay of the thermal response, a slope of the thermal response, a maximum amplitude of the thermal response, or a relaxation time of the thermal response;
and inferring an estimate of the state of health of the electrochemical device from the measurements of the parameters. The method of claim 1, wherein the measured parameter includes a delay (tdelay) until the thermal response of the electrochemical device begins relative to the start time of the application of the heat supply. The method of claim 1, wherein the heat supply is applied continuously after a start time of the application of the heat supply, the thermal response of the electrochemical device is recorded as a function of time, and the measured parameter comprises the slope of variation of the thermal response as a function of time (dP/dt). The method of claim 1, wherein the heat supply is applied continuously for a selected duration after a start time of the application of the heat supply, and then stopped after the selected duration, and the measured parameter comprises the maximum amplitude (Pmax) of the thermal response of the electrochemical device. The method of claim 1, wherein the measured parameter includes the length of time (τ) for the thermal response of the electrochemical device to return to a predefined threshold value after terminating the application of the heat supply. The method of claim 1 , wherein the electrochemical device is in operation during the application of the heat supply. The method of claim 1, wherein the parameter measurements are stored in memory to monitor changes in the state of health of the electrochemical device over time. The method of claim 7 , comprising generating a warning signal when the state of health of the electrochemical device is below a threshold value. 1. A device for estimating the state of health (SoH) of an electrochemical device, comprising:
at least one sensor for detecting a thermal response of the electrochemical device to a heat supply applied to the electrochemical device;
a processing circuit for measuring at least one parameter indicative of a thermal inertia of the electrochemical device in the thermal response, the thermal inertia being defined as the time it takes for the electrochemical device to change temperature in response to the supply of heat, the thermal inertia of the electrochemical device increasing as the State of Health (SOH) of the electrochemical device decreases, and the parameter being selected from a delay of the thermal response, a slope of the thermal response, a maximum amplitude of the thermal response, or a relaxation time of the thermal response; and for inferring an estimate of the State of Health of the electrochemical device from the measurement of the parameter. The device of claim 9 , further comprising a heat transfer device attached to the electrochemical device and configured to apply the heat supply to the electrochemical device. The device of claim 10 , further comprising a pad made of an electrically insulating material, the heat transfer device being configured to be attached to the first major surface of the electrochemical device by the pad. The device of claim 9 further comprising a heat sink attached to the electrochemical device. 10. The device of claim 9 , wherein the processing circuitry comprises a memory for storing measurements of at least some of the parameters and for monitoring changes in the state of health of the electrochemical device over time. A computer program comprising instructions which, when executed by a processing circuit, implement the steps of the method of any one of claims 1 to 8 .