Method for estimating consumed life of energy storage device

Abstract

The invention relates to the field of energy storage devices, in particular for estimating the consumed life of an energy storage device. A method of estimating consumed life of an energy storage device includes: measuring a time series of ambient relative humidity and associated temperature of the energy storage device; a consumed life of the energy storage device is estimated based on the time series and a moisture diffusion model of the energy storage device.

Description

Technical Field

[0001] The present invention relates to the field of energy storage devices, in particular to the field for estimating the consumed life of energy storage devices. The present invention also relates to a non-transitory computer-readable storage medium, an apparatus, and to a use. Background Art

[0002] There may be many reasons for the decay and / or shortening of the life of an energy storage device. One of the reasons may be humidity, which diffuses into the energy storage device in the form of moisture. The end of life may be harmful to the device and / or even to people, because the moisture inside may even cause an explosion, especially if the diffused moisture (i.e., water) is decomposed into hydrogen and oxygen gases. Due to this risk, at least in some cases, energy storage devices are properly disposed of (sorted out) before the actual end of their life, otherwise they will run the risk of being properly disposed of too late. Therefore, components for estimating the consumed life of energy storage devices may be useful. Summary of the invention

[0003] It is an object of the invention to provide a means for estimating the consumed life of an energy storage device. This object is achieved by the subject matter of the independent claims. Further embodiments become apparent from the dependent claims and the following description.

[0004] One aspect relates to a method for estimating the consumed life of an energy storage device, the method comprising the steps of:

[0005] Measuring a time series of ambient relative humidity and associated temperature of the energy storage device; and

[0006] The consumed life of an energy storage device is estimated based on a time series and a moisture diffusion model of the energy storage device.

[0007] Estimating the consumed life of an energy storage device can be, for example, an indicator for predicting the life (or end of life) of an energy storage device and / or for estimating the current health state of the device. In order to estimate the consumed life of an individual energy storage device, a time series of relative humidity and a related temperature of the environment of the energy storage device is measured. The humidity measurement itself can be based on the relative humidity and / or absolute humidity of its environment. Therefore, it is foreseeable that a means of converting it into a desired measurement scale can be used. The relevant temperature of the energy storage device can be measured directly at the energy storage device (at a certain distance from the energy storage device, preferably close to the energy storage device, and / or for some energy storage devices, it may be inside the device). In addition or as an alternative, the temperature can be calculated based on some measured temperatures and a thermal model of the energy storage device and / or its surroundings. For example, if the ambient temperature measurement is the only measurement available, the internal temperature can be obtained by adding a temperature difference that takes into account ohmic heating. For example, the value can be determined by knowing the electrical parameters of the capacitor and measuring the current. If the current cannot be measured, the ambient temperature can be corrected by a term determined by a previous understanding of the behavior of the capacitor. In case the method is applied to several capacitors in a bank and only one ambient temperature is available, the influence of each capacitor position on the thermal behavior can be taken into account.The measurements can be made regularly (ie, periodically), continuously and / or on demand.

[0008] In other words, it is known that at least some types of energy storage devices, such as some batteries or metallized film capacitors (MFCs), are affected by ambient humidity. Some capacitors can be particularly dangerous when operated at AC voltage. Typically, as soon as moisture penetrates the housing and reaches the metallized film inside, electrochemical corrosion and / or corona discharges occur, leading to degradation of the membrane. This, in turn, is reflected in changes in electrical parameters, such as the capacitive and ohmic part of the impedance (ESR). Ultimately, the capacitor may fail catastrophically. To prevent this, the method provides a means for monitoring its condition and / or predicting its future behavior, in particular estimating its consumed life.

[0009] Methods based on direct measurements of these electrical parameters to observe the progression of degradation may be technically difficult to implement and difficult to interpret. In particular, these direct measurements may be inappropriate or inconvenient because the electrical parameters of the energy storage device may not change during the period before moisture reaches the metal film. After that, the degradation rate of the capacitor may be too fast to be predicted in time, and the remaining life of the capacitor may also be difficult to estimate because the further development of damage depends on the entry of water into the metal film.

[0010] One way to deal with or at least mitigate this problem is to use humidity measurement. Directly sensing the (total) moisture content absorption in the capacitor via measuring the mass increase may be difficult and may not provide information about the spatial distribution of the water inside the capacitor. Measuring the ambient humidity is easy, but needs to be correlated with the consumed life over time. The standard way to relate external humidity to consumed life is to use the so-called Peck-Hallberg model, which is commonly used for electronic devices placed inside electronic packages (for example, inside filled plastic housings). It provides an acceleration factor that depends only on the ratio between the instantaneous relative humidity and the reference nominal humidity and the temperature in the Arrhenius form. However, the Peck-Hallberg model is purely empirical and has two main drawbacks:

[0011] 1) The model is only applicable to cases of constant conditions. When applying it to variable conditions in the normal way, it is assumed that humidity has an instantaneous effect. However, experiments have shown that there is a retardation effect due to the physical barrier that the water must pass through before reaching the metal film. Since the Peck-Hallberg model is generally only applicable to cases of constant environmental conditions, this retardation can be neglected. For changing conditions, the model may not be adequate.

[0012] 2) This means that the consumed life will only increase with time. However, high temperatures and / or low concentrations of humidity may dry out the capacitor by diffusing and releasing moisture to the environment, which may slow down, delay, or even stop the degradation process and may also reduce the internal moisture concentration. Therefore, under such conditions, the consumed life should be able to be reduced or restored.

[0013] Thus, the method is based on a physics-inspired model that is based on temperature and relative humidity measurements to track the diffusion of moisture inside an energy storage device or MFC and thereby determine its lifetime. The method described above and / or below can be used to determine the lifetime of a single capacitor or several capacitors in a group of capacitors. Thus, based on a method that takes into account the time series of humidity and temperature and based on a moisture diffusion model for the energy storage device, the consumed lifetime of the energy storage device can be estimated. The moisture diffusion model (or hygroscopicity, humidity, water diffusion model) can be based on the amount of water in the energy storage device, which is calculated from the measured time series and the selected model.

[0014] Advantageously, the method no longer needs to rely on standard assumptions about the environment in which the energy storage devices operate, but instead bases its estimates on the actual environmental conditions of each energy storage device. This can prevent both the risk of (potentially harmful) damage (for example, when the device stays in its system for too long) and the unnecessary premature end of the system life. The estimated consumed life can be used as an indicator for predicting the end of the life of the energy storage device and / or for estimating the current health state of the device. The estimated consumed life can be the basis for outputting an alarm, a maintenance request, and / or for requesting and / or for initiating countermeasures (for example, such as means for drying the energy storage device), and / or for another active and / or passive measure.

[0015] In various embodiments, the moisture diffusion model takes into account the spatial distribution of moisture inside the energy storage device. Therefore, the moisture concentration and the spatial distribution of moisture inside the energy storage device are taken into account in the model.

[0016] In various embodiments, the moisture diffusion model accounts for the diffusion of moisture into and within the energy storage device.

[0017] In various embodiments, the moisture diffusion model takes into account the temperature inside the energy storage device, and / or the moisture diffusion model estimates the temperature inside the energy storage device. For example, the estimated temperature inside the energy storage device can be based on the distance from the energy storage device and / or the heat transfer resistance of the device.

[0018] In various embodiments, the parameters of the water diffusion model are obtained by performing aging tests on specific conditions (e.g., constant conditions). These data can be used to determine the relevant parameters of the model, for example, to be consistent with or as close as possible to the test parameters.

[0019] In various embodiments, the consumed life is a function of the amount of water present in the interior portion of the energy storage device. The function may be based on one or more models of the energy storage device (see below).

[0020] In various embodiments, the consumed life and / or water volume is stored as a single variable or a distribution function. The distribution function can be stored as an array, as a mathematical function, and / or in another form. The model or distribution function can take into account the metal film in the internal portion of the energy storage device.

[0021] In an embodiment, the distribution function is based on a 3D, 2D or 1D model of the energy storage device, taking into account 3 dimensions, 2 dimensions or 1 dimension of the device. Where a 3D model may be more accurate, a 2D model or even a 1D model may be easier to apply, resulting in a model that requires less computing power to perform the estimation.

[0022] In order to estimate the consumed life, a model can be solved that simulates the diffusion of moisture into and within the energy storage device. In general, the model can be in the form of a partial differential equation for diffusion in a medium, possibly with one or more interfaces. For example, the first interface can be an interface with the outside air, and the last interface can be an interface connected to a region (corresponding to the body of the energy storage device) in which moisture causes degradation (e.g., corrosion). The boundary conditions (i.e., the flux of moisture into the medium) can depend on the value of the measured relative humidity. In addition or as an alternative, more complex reaction-diffusion models can be implemented, for example, to describe the corrosion of metal films with water, or to permanently bond to organic ("plastic") materials in the housing, and to represent damage in this way. The output of the model can be a function that describes and / or summarizes the spatial distribution of moisture inside the energy storage device.

[0023] In order to solve the model, a computing unit is useful or necessary, which may be connected to a cloud and may require communication means (e.g., a LAN or any other type of connection). The unit may have measurements of temperature and relative humidity as input and may have access to a storage unit where the solution to the model is stored (e.g., applied to one or more previous time steps).

[0024] Based on this, the solution of the model (i.e., the distribution of moisture within the medium) can be converted into the consumed life. In some embodiments, the consumed life can be stored as a single variable that can be updated with each calculation. Storing a single variable is an efficient way to process the results of the calculation. This conversion function can be selected in such a way that both the moisture concentration (in the medium or for corroding some materials) and the spatial distribution of moisture inside the energy storage device are taken into account. For example, the conversion utilizes the amount of water present in the energy storage device and how much water has reached the inside of the metal film. The moisture concentration can then be associated with a reference value that corresponds to how much moisture can reach the internal area before the energy storage device reaches the end of its life. Taking the ratio of the two, when the value of the moisture content reaches the reference value, the consumed life will be equal to 1. According to the definition used to indicate a high risk of failure device and / or according to requirements, a value greater than 1 can be reached.

[0025] A specific example is provided for a deeper understanding. The example is based on the one-dimensional diffusion equation

[0026]

[0027] In the simplest version, the domain for this equation can be considered semi-infinite. The boundary condition at x=0, corresponding to the interface with the external environment, can be obtained by converting the (ambient) measurement of the relative humidity to the relative humidity at the surface and inside the capacitor (e.g., correcting for different temperatures via the Antoine equation). The diffusion constant D depends on the temperature according to the Arrhenius law, and its parameters depend on the material properties of the encapsulation or housing material (which can be determined in advance and used in the model).

[0028] This equation can be solved to obtain the concentration C(x,t) using a suitable time integration scheme well known in the relevant literature and discretizing the spatial domain accordingly. Other methods, for example, those based on Green's functions, can also be used.

[0029] The conversion to consumed lifetime can be obtained by integrating the solution C(x, t) from a certain position, such as x=1 (the boundary of the simulated metal film region) to infinity. The consumed lifetime will be the ratio between the result of this integration and the reference value.

[0030] Note that the model contains only a small number of parameters, which allows them to be set based on experimental results (e.g. using constant or stepwise constant conditions). In this way, the parameters are not necessarily identical to the physical properties, but reflect the lifetime observed under these conditions (e.g. from ALT (accelerated lifetime) tests).

[0031] In an optional step, the consumed life can be further transformed: the total moisture content in a region may not depend linearly on the consumed life itself. However, by simulating constant reference conditions, a reference curve of total moisture content versus consumed life can be calculated. Inverting this curve and using quantities derived from the model under variable conditions within the model can yield a consumed life value that will be linear under constant conditions and therefore easier to interpret.

[0032] In some embodiments, the result "current consumed life" can be treated as a single variable that is used as the basis for the following calculations. Advantageously, this can save some storage space, because only this variable can be saved in persistent memory (e.g., in long-term persistent memory and / or in special registers of the system in which the (multiple) energy storage devices are placed), and only the time series of measurements of temperature and relative humidity can be used for calculations. After the calculations, this time series can be deleted.

[0033] In some embodiments, each measurement of the time series is performed on demand and / or periodically, with a period of one minute, one hour, or one day. The frequency of measurement may depend on the system in which the energy storage device is arranged. For example, measurements in a portable system (e.g., a laptop) may be performed more frequently than measurements in an air conditioning system (e.g., a server).

[0034] In some embodiments, the time series of measurements is stored for a week, a month, or a year.

[0035] In some embodiments, the energy storage device is a battery. Examples may include batteries such as Li-Ion (lithium ion), LTO (lithium titanate), LMO (manganese), LiFP (lithium iron phosphate), nickel-based (such as NiMH (nickel metal hydride) or NiCd (nickel cadmium) batteries). In general, the methods and models described above and / or below can be applied to any battery whose life is determined and / or affected by the amount of water diffused into the battery.

[0036] In some embodiments, the energy storage device is a capacitor, in particular a film capacitor, the film comprising polypropylene (PP), polyethylene terephthalate (PET), polyethylene naphthalate (PEN), polyphenylene sulfide (PPS), polycarbonate (PC), functionally enhanced sub-variants including nanostructured materials, additional polymers and / or combinations of these materials. For example, a polymer film type capacitor can be used as a filter capacitor (its capacity is between 1 μF and 5000 μF, in particular between 10 μF and 100 μF) and is specified for voltages between 100 V and 3000 V (e.g., 350 V AC). The edge length of such a capacitor is between 10 mm and 500 mm, in particular between 20 mm and 60 mm. The capacitor type may include an electrolytic capacitor type.

[0037] In some embodiments, the capacitor includes at least one internal barrier. The internal barrier can be implemented, for example, as a thin layer or a thicker (i.e., more extended) region. For example, the internal barrier can be an epoxy resin or any other moisture-proof polymer, alternatively a thin metal or other material layer that has a higher moisture diffusion barrier behavior than the main fill material. The internal barrier can slow down the intrusion of moisture, thereby extending the life of the energy storage device. The diffusion model can combine the effects of one or more internal barriers and / or the effects of chemical reactions of water with the metal and / or other materials of the internal barriers in the energy storage device.

[0038] One aspect relates to a non-transitory computer-readable storage medium having stored therein a program which, when executed on a computing unit, instructs the computing unit to perform the method described above and / or below.

[0039] One aspect relates to an apparatus for estimating the consumed life of an energy storage device. The apparatus comprises: a humidity sensor configured to measure the relative humidity of the environment of the energy storage device; a temperature sensor configured to measure the temperature of the energy storage device; a memory configured to store a time series of humidity and temperature measurements of the energy storage device and / or the amount of water present in the internal part of the energy storage device; and a computing unit configured to estimate the consumed life of the energy storage device according to the method described above and / or below and optionally store the value of the estimated consumed life of the energy storage device.

[0040] The temperature sensor may be arranged on the surface of the energy storage device and / or arranged inside a device containing the energy storage device. In some embodiments, only one temperature sensor is used to determine the temperature of multiple energy storage devices.

[0041] One aspect relates to using the apparatus described above and / or below to estimate the consumed life of an energy storage device.

[0042] It should be noted that two or more embodiments described above and / or below may be combined as long as technically feasible.

[0043] For further explanation, the present invention is described with the aid of the embodiments shown in the drawings. These embodiments should be regarded as examples only and not as limitations. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings depict:

[0045] Figure 1a , Figure 1b , Figure 1c schematically illustrates an energy storage device according to an embodiment;

[0046] Figure 2a , Figure 2b , Figure 2c schematically illustrates an energy storage device according to an embodiment;

[0047] Figure 3a , Figure 3b shows a solution to a diffusion model according to an embodiment;

[0048] Figure 4 is an example of the consumed life of an energy storage device according to an embodiment;

[0049] Figure 5 is a flow chart according to an embodiment. DETAILED DESCRIPTION

[0050] Figure 1a , Figure 1b , Figure 1cEnergy storage devices according to embodiments are schematically illustrated. In these figures, capacitors are used as examples, but this information can also be easily applied to other types of energy storage devices. These figures will depict how the moisture concentration inside the capacitor (depicted as rectangles) is at least partially reversible in some cases. The water droplets within the frame schematically illustrate the moisture concentration levels inside the capacitor and their spatial distribution over time. Figure 1a A starting point is shown on the left with a "dry" capacitor, i.e., a capacitor with no water inside (e.g., a new capacitor). After being exposed to a high ambient relative humidity level ("Level 1", on the left, shown with 5 drops), water will first accumulate in the outer areas, indicating that some water has penetrated, but has not yet damaged the capacitor because the water has not yet reached the core; see Figure 1a In the middle, it shows a moderate humidity distribution ("spatial distribution 1"), i.e., an increase in humidity levels near the capacitor border. After further (e.g., prolonged) exposure to "ambient humidity level 1", water may have reached the interior as well; see Figure 1a On the right, it shows a high humidity concentration ("Spatial Distribution 2"), which is almost uniformly distributed inside the capacitor. If "Spatial Distribution 2" occurs, the capacitor will be damaged and the estimated consumed life will be 100%.

[0051] Figure 1b The effect of a lower ambient relative humidity level ("Level 2", shown with 2 drops) is shown. This may result in a "reversal" of the estimated consumed lifetime. From "Spatial Distribution 1" (see Figure 1a Middle and Figure 1b The capacitors may then be moved to a location with "ambient humidity level 2" (i.e., the ambient relative humidity level is quite low and / or countermeasures may be taken, for example, by placing the capacitors in a dryer). Figure 1b As shown in the middle, the resulting effect ("spatial distribution 3") may be twofold: on the one hand, water continues to penetrate, i.e., some water will reach the interior; on the other hand, the amount of water decreases, resulting in a decrease in the estimated consumed life. After further exposure to "ambient humidity level 2", the estimated consumed life continues to decrease, resulting in "spatial distribution 4"; see Figure 1b In this example, moisture eventually reaches the interior, but the humidity level is much lower, resulting in a lower estimated life consumed. Figure 1b It is clearly shown that exposure history can be important and that removing moisture makes the lifetime reversible.

[0052] Figure 1cThe effect of increasing the ambient relative humidity level again on the spatial distribution of moisture inside the capacitor is shown, i.e. the moisture concentration slowly increases over time. In this example, the capacitor is exposed to a high ambient relative humidity level (“Level 3”, shown with 5 drops compared to “Level 1”). Starting from “Spatial Distribution 1” (see Figure 1b Right and Figure 1c left side), i.e., the capacitor is exposed at "Level 3". As a result, "Spatial Distribution 5" appears (see Figure 1c Middle): The water inside the capacitor remains unchanged at first, but the moisture in the outer area increases. Although this effect is not immediate, the moisture first stays in the outer area and then increases further, eventually causing the capacitor to fail; see Figure 1c The "Spatial Distribution 6" on the right is similar to Figure 1a "Spatial Distribution 2" on the right.

[0053] Figure 2a , Figure 2b , Figure 2c Schematically shows an energy storage device according to an embodiment. The energy storage device in these figures has an additional internal barrier. The situation of increasing and decreasing relative humidity of the environment is similar to Figure 1a , Figure 1b , Figure 1c It can be clearly seen that the moisture content inside the device also increases and decreases depending on the ambient relative humidity level. However, Figure 2a From "dry" ( Figure 2a Left) to "Spatial Distribution 1" (see Figure 2a The time from “spatial distribution 1” to “spatial distribution 2” ( Figure 2a The same is true for the time period of the right side). Therefore, high external humidity "Level 1" first causes accumulation in the outer layer (shown here to the barrier). The barrier is therefore intended to prevent or at least delay the moisture content from entering further into the interior for a period of time. This effect is similar to Figure 2b and Figure 2c .therefore, Figure 2a , Figure 2b , Figure 2c It is explained why it is advantageous to consider an additional internal barrier inside the capacitor.

[0054] Figure 3a , Figure 3b The solution of a diffusion model according to an embodiment is shown. The x-axis shows the position x within the device. The y-axis shows the water density or concentration rh ("relative humidity", ie relative to the saturation density in the surrounding air; it is therefore usually normalized). Figure 3aThe solution of a fairly realistic version of the diffusion model after one year of exposure using some real meteorological conditions (e.g., Singapore's) is shown. The curve is plotted weekly, and its progression reflects the increasing darkness later on. Figure 3b The solution is shown over one year at constant conditions T = 20°C and external relative humidity = 0.4. These figures generally show the humidity or water distribution inside the energy storage device or its model, which can be very complex and depends on the humidity and temperature history.

[0055] Figure 4 An example of the consumed life of an energy storage device according to an embodiment is shown. The y-axis shows a value representing the progression of water in the model, where the distance to the interior is and where the total content reaches a critical amount. This can be used as an estimate of the consumed life x c =CL basis. The x-axis shows the exposure time of the energy storage device to two different ambient relative humidity levels applied to the energy storage device in two different time spans. The dotted line separates the two halves of the time span. In the first half of the time span, the energy storage device is placed under a high ambient relative humidity level (e.g., Figure 1a During the second half of this time span, the energy storage device is placed in a low ambient relative humidity level (e.g., Figure 1b This leads to the following effect: At first (e.g. until day 200) the internal humidity inside the device remains low enough not to reach critical values. This is followed by an almost linear increase. When placed under low ambient relative humidity levels, the internal humidity still increases for a few days and then starts to decrease. After a while, this decrease resumes again due to further diffusion into the device.

[0056] Figure 5 A flow chart 100 is shown according to an embodiment, reflecting a method for estimating the consumed life of an energy storage device. In step 102, a time series of ambient relative humidity and a related temperature of the energy storage device is measured. In step 104, the consumed life of the energy storage device is estimated based on the time series and a humidity diffusion model of the energy storage device.

Claims

1. A method for estimating the consumed life of an energy storage device, the method comprising the steps of: measuring a time series of ambient relative humidity and associated temperature of said energy storage device; as well as An consumed life of the energy storage device is estimated based on the time series and a moisture diffusion model of the energy storage device.

2. The method according to claim 1, The moisture diffusion model takes into account the spatial distribution of moisture inside the energy storage device.

3. The method according to claim 1 or 2, The moisture diffusion model takes into account the diffusion of moisture into the energy storage device and within the energy storage device.

4. The method according to any one of the preceding claims, wherein the moisture diffusion model takes into account the temperature inside the energy storage device, and / or The moisture diffusion model estimates the temperature inside the energy storage device, and / or The parameters of the moisture diffusion model are obtained by performing aging tests on specific conditions.

5. The method according to any one of the preceding claims, Wherein the consumed life is a function of an amount of water present in an interior portion of the energy storage device.

6. The method according to any one of the preceding claims, The consumed lifespan and / or the amount of water are stored as a single variable or a distribution function.

7. The method according to any one of the preceding claims, The distribution function is based on a 3D, 2D or 1D model of the energy storage device.

8. The method according to any one of the preceding claims, wherein each measurement of the time series is performed on demand and / or periodically, The period is one minute, one hour or one day.

9. The method according to any one of the preceding claims, The time series of measurements are stored for one week, one month or one year.

10. The method according to any one of the preceding claims, The energy storage device (10) is a capacitor, in particular a membrane capacitor, wherein the membrane comprises polypropylene (PP), polyethylene terephthalate (PET), polyethylene naphthalate (PEN), polyphenylene sulfide (PPS), polycarbonate (PC), functionally enhanced sub-variants including nanostructured materials, other polymers and / or combinations of these materials.

11. The method according to claim 10, Wherein the capacitor comprises at least one internal barrier. 12 . A non-transitory computer-readable storage medium having a program stored therein, which, when executed on a computing unit, instructs the computing unit to perform the method according to claim 1 .

13. An apparatus for estimating the consumed life of an energy storage device, the apparatus comprising: a humidity sensor configured to measure a relative humidity of an environment of the energy storage device; a temperature sensor configured to measure a temperature of the energy storage device; a memory configured to store a time series of humidity and temperature measurements of the energy storage device and / or an amount of water present in an interior portion of the energy storage device; as well as A calculation unit configured to estimate the consumed life of the energy storage device according to the method of any of the preceding claims and optionally store a value for estimating the consumed life of the energy storage device.

14. Use of the apparatus according to claim 13 for estimating the consumed life of an energy storage device.

Images