Method for determining aging of individual battery cells

Abstract

The invention relates to a method for determining the aging of such an individual cell in a lithium-ion battery consisting of a plurality of cells, for which purpose a cell thickness change (x) of the individual cell with respect to the state of charge (SOC) is detected in order to derive therefrom a conclusion relating to the aging of the individual cell. The method according to the invention is characterized in that the curve of the cell thickness variation (x) is evaluated over the entire state of charge (SOC) during charging and discharging of the individual cell in such a way that the characteristic properties of the curve are used to infer the thickness variation (x) of the active materials (Si, Gr, Si). NMC) with respect to the cell thickness variation (x) and thus deducing the degree of engagement with the aging.

Description

[0001] The present invention relates to a method for determining the aging of a single battery cell, the type of which is defined in more detail in the preamble of claim 1.

[0002] In principle, it is known from existing technology that the volume of a single battery cell increases with increasing charge and / or decreases with increasing discharge, which is particularly true for lithium-ion batteries. Furthermore, there is a situation where the volume also changes as a single battery cell ages.

[0003] DE 10 2021 005 418 A1 utilizes this to determine the aging state of an individual battery cell using a cell thickness-based method. Here, the cell thickness variation of an individual battery cell with respect to its state of charge is detected, and this variation is then divided into a portion caused by charging or discharging and a portion caused by the aging of the individual battery cell. This allows for the deduction of the aging of the individual battery cell.

[0004] Here, we now find that the changes in cell thickness related to aging exhibit different characteristics depending on the material. Hendrik Pegel et al., in their article "Volume and thickness change of NMC811|SiO2" published in the Journal of Power Sources 537(2022)231443, address this issue. x The study, "Graphite Large-Format Lithium-ion Cells: From Pouch Cell to Active Material Level," demonstrates the correlation in this area. Its core research indicates that the overall cell thickness variation during charge-discharge cycles is partly composed of cell thickness variations caused by silicon and graphite (anode side) and partly by manganese-nickel-cobalt materials known as NMC (cathode side). In other words, the sum of cell thickness variations is the sum of cathode aging and anode aging.

[0005] The object of the present invention is now to create an improved method for determining the aging of individual battery cells compared with the prior art, in which the aging caused by the corresponding active materials can be deduced.

[0006] According to the invention, this objective is now achieved by a method having the features of claim 1, particularly the features in the characterizing portion of claim 1. Advantageous designs and improvements are given in the dependent claims.

[0007] Similar to the prior art mentioned at the beginning, the cell thickness change of an individual battery cell as a whole is detected here. This cell thickness change, relating to the state of charge, and particularly to the charge and discharge cycles of an individual battery cell, then yields a corresponding curve for the cell thickness change. This evaluation of each individual battery cell allows for the inference, using characterizing properties, of the degree to which at least one active material contributes to the cell thickness change, and thus ultimately the inference of the aging of that active material. The principle of the characterizing properties of the individual active material for the cell thickness change itself is known from the aforementioned scientific publications. Now, using these characterizing properties, the degree to which the active material used in each case contributes to the cell thickness change is deduced from the detected aging-related cell thickness change. What ultimately becomes feasible is to determine the aging of that one active material, i.e., the deintercalation of graphite or silicon in the anode region or the loss of active material in the cathode region, for example, caused by a deposition process or similar process.

[0008] Therefore, when controlling batteries, it is possible to purposefully omit or adjust operating conditions that have a particularly strong impact on the aging of one or more active materials. This makes it possible to efficiently halt the aging process.

[0009] Furthermore, this method makes it possible to decompose the effect of individual active materials on cell thickness variations by simply measuring cell thickness changes. This allows for the relatively easy collection of data under real-world operating conditions in fleets of electric or hybrid vehicles, which can then be used to deepen the understanding of the aging process of such individual battery cells.

[0010] That is, by performing a curve evaluation on the characterization characteristics of at least one active material according to the present invention, it is in principle possible to infer the portion of the cell thickness variation that occurs throughout the entire lifespan of a single battery cell by that active material. According to a highly advantageous improvement, the above-mentioned curve evaluation is performed correspondingly for all active materials (i.e., silicon, graphite, NMC, lithium metal, and lithium iron phosphate), with the premise that there is a linear relationship between the volume fraction of the corresponding active material and its effect on the cell thickness variation.

[0011] A particularly advantageous design of the method according to the invention is to assess the plateau hysteresis of cell thickness variation between the charging and discharging plateaus in order to determine silicon aging. The curve of cell thickness variation of the entire single cell will always have a characteristic plateau during charging and an upper characteristic plateau during discharging. The distance between these two plateaus is referred to here as the plateau hysteresis. This plateau hysteresis can now be assessed so that the silicon-induced portion of the cell thickness variation can be inferred from the measured cell thickness variation of the entire single cell. In particular, studies have shown that the plateau hysteresis at the beginning of the lifespan of a single cell is significantly greater than the plateau hysteresis when the single cell has aged. That is, silicon aging can be deduced from the magnitude of the plateau hysteresis. Studies have shown that silicon aging determined by means of plateau hysteresis in the cell thickness variation curve characterizing silicon is consistent with silicon aging determined by means of the cell thickness variation curve of the entire cell. That is, silicon aging can be determined correspondingly by means of two thickness values ​​of the first and second plateaus that are relatively related to each other.

[0012] Another highly advantageous design approach involves evaluating the height of the final charging platform based on the cell thickness variation to determine graphite aging. Here, the height of the final charging platform also corresponds to the cell thickness variation caused by graphite aging. Specifically, if silicon aging is already known and the silicon portion of the cell thickness variation in the final charging platform is therefore also known, the graphite-related portion of the final charging platform height can be easily deduced from the difference, allowing for the corresponding calculation of graphite aging. Here, with silicon aging known, NMC aging has almost no effect in the region of the second final charging platform, so its portion can be ignored without loss of accuracy.

[0013] Then, to determine NMC aging, the maximum value of the cell thickness variation with respect to the discharge state can be evaluated. Typically, this maximum value will be located in the region where the cell is fully charged, i.e., in the region of 100% state of charge. The maximum value appearing in this region then allows for the deduction of the characterizing minimum value that typically appears in this region in the cell thickness variation caused by NMC. Given that silicon aging and graphite aging are known, NMC aging can therefore be easily determined.

[0014] Therefore, a highly advantageous improvement according to the invention is set up by first determining silicon aging, then graphite aging, and finally NMC aging by characterizing the cell thickness variation of the corresponding active materials. The advantage of this is that the known plateau height of the cell thickness variation caused by silicon can already be included when determining graphite aging, and the other two values ​​can already be taken into account when determining the magnitude of the minimum value of the cell thickness variation related to NMC.

[0015] Now, in order to determine this over the entire relevant portion, particularly over the entire state of charge (SOC), a highly advantageous improvement to the method according to the invention proposes to use the value of the silicon-related cell thickness change with respect to SOC determined from the plateau hysteresis, to compress the known initial curve of the silicon-related cell thickness change with respect to SOC at the start of the lifetime in the direction of cell thickness change, so as to obtain a silicon-related cell thickness change curve reflecting the current aging degree over the entire SOC during the charge-discharge process. The measured value of the plateau hysteresis also allows for the derivation of the cell thickness change occurring in the corresponding plateau, thereby making it possible to determine the height of the plateau in the curve accordingly. Thus, it is now possible to calculate the current curve of the silicon-related cell thickness change based on the known curve of the silicon-related cell thickness change at the start of the cell lifetime, by compressing the curve according to the determined value. Therefore, this current curve represents the value of the silicon-related cell thickness change over the entire charge-discharge period in the current aging state of a single cell. This curve can then be subtracted from the measured total curve to obtain the portion of the graphite and NMC-related cell thickness change in the current aging state. To determine this, regions characterizing the corresponding materials are now used, particularly in the sense described above, to determine graphite-related cell thickness variations and thus ultimately graphite aging, using the height of the final charging platform.

[0016] Similar to using values ​​detected from the plateau hysteresis in the case of silicon-related cell thickness variations, it is now possible to compress the known curves for the graphite-related portion of the cell thickness variation at the start of the life of a single cell based on the height of the discharge plateau in the case of graphite-related cell thickness variations, so that in addition to obtaining the silicon-related cell thickness variation reflecting the current aging condition, a graphite-related cell thickness variation reflecting the current aging condition is also obtained as a new curve for the current aging state of a single cell.

[0017] Then, by specifically using these two curves, it is possible to deduce the portion of the cell thickness variation caused by NMC that accounts for the total cell thickness variation, so as to identify the peak value in the minimum of the NMC curve in such a simple and efficient way, and detect the NMC-related cell thickness variation from its magnitude and thus detect the aging state of NMC, which is essentially the aging state of the cathode.

[0018] Specifically, according to a favorable design scheme for determining silicon aging, the difference in cell thickness between the charging plateau and the final charging plateau of the curve relating cell thickness to the state of charge can be evaluated. A decrease in the distance between the plateaus, i.e., a smaller plateau hysteresis, here signifies accelerated silicon aging.

[0019] According to an advantageous improvement, to determine graphite aging, the graphite-related height of the final charging platform can be determined from the difference between the height detected in the cell thickness variation curve of the entire single battery cell on the final charging platform and the previously determined silicon-related height variation of the final charging platform. Then, by using the magnitude or height of the decrease in the graphite-related cell thickness variation of the final charging platform, accelerated graphite aging can be inferred. Furthermore, using the two determined aging methods, it can also be set up to determine the corresponding maximum values ​​of the battery thickness variation caused by silicon and graphite in the fully charged state, and determine the minimum value of the cell thickness variation caused by NMC in the fully charged state from the difference between the total maximum value and the graphite and silicon-related maximum values, wherein the magnitude of the decrease in this minimum value is used to infer accelerated NMC aging.

[0020] That is, by simply evaluating the corresponding features in the curves, it is possible to easily and efficiently determine the distribution of aging-related cell thickness variations among the different active materials involved, i.e., their impact on the corresponding magnitude of cell thickness variations. As mentioned above, this can be used to improve control or operating strategies for such batteries, as well as for extensive data collection to better analyze and understand the aging process in such individual battery cells.

[0021] The method of the present invention will be described in detail below through embodiments, with reference to the following figures.

[0022] in:

[0023] Figure 1 A schematic diagram illustrating an embodiment of the method of the present invention is shown.

[0024] Figure 2 A schematic diagram showing the percentage expansion of a single battery cell (expressed as a percentage of state of charge) is shown, in which curves are plotted for the entire cell, the anode (made primarily of silicon and graphite), and the cathode (made of NMC).

[0025] Figure 3 Two curves showing the percentage expansion (expressed as a percentage of cell thickness) are shown, representing the cell thickness change with respect to the state of charge at the beginning of the life of a single cell (a) and in the aged state (b).

[0026] Figure 4 according to Figure 3 The silicon-related portion showing the variation in cell thickness is illustrated.

[0027] Figure 5 Similar to Figure 3 and Figure 4 The graphite-related portion showing the variation in unit thickness is shown; and

[0028] Figure 6 Similar to Figure 3 , Figure 4 and Figure 5 Alternatively, these figures may be supplemented to show the NMC-related portion of the element thickness variation.

[0029] exist Figure 1 The diagram illustrates the implementation of the proposed method in the form of a flowchart. The first square, marked 1 on the far left, represents a dynamic in-operando measurement within the vehicle, in which the change in the expansion of an individual battery cell is detected as a function of the state of charge. This is also referred to below as cell thickness variation, which essentially represents the percentage change in thickness of an individual battery cell with respect to its state of charge (from an empty single battery cell to a fully charged state, and then to re-discharge).

[0030] Figure 2 The middle square, marked 2 in the diagram, essentially represents the battery control unit 2. Starting from the dynamic in-situ measurement 1 within the vehicle, the measured cell thickness variation with respect to the state of charge is transmitted to the battery control unit 2. Then, in the first square 2a, theoretical knowledge of dividing the cell thickness variation into portions caused by the corresponding active materials is applied to establish a correlation between cell thickness variation and active material aging throughout the aging process of a single battery cell. This is represented by the square marked 2b. Subsequently, the corresponding target information 3 is used again, either within the battery control unit itself to adjust control in a way that minimizes the effects of aging, or to collect data that helps in better understanding aging.

[0031] exist Figure 2The diagram now shows the percentage expansion of cell thickness (expressed as a percentage of state of charge), referred to below as cell thickness variation. This curve reflects the initial state of a single battery cell at the start of its lifespan. Cell thickness variation here varies between -1% and approximately +5.5%. The state of charge is fully charged at the midpoint, so discharge is shown to the left of the midpoint, and charging is shown to the right, i.e., charging occurs to the left of the midpoint as it ranges from 0% to 100% SOC on the X-axis, while discharging occurs to the right of the midpoint as it ranges from 100% SOC to 0%. The cell thickness variation measured dynamically in-situ is represented by a bold curve with a solid line. First, when dividing the cell thickness variation into individual portions, it is divided into a cathode portion, represented by a dashed line, and an anode portion, represented by a double dashed line. The portion of the cathode-related cell thickness variation x with respect to state of charge (SOC) here essentially corresponds directly to NMC-related aging, as NMC is the base material of the cathode. The thickness variation of the anode-related unit is further divided into two active materials of the anode: silicon and graphite. The thickness variation of the silicon-related unit is represented by a dashed line, and the thickness variation of the graphite-related unit is represented by a dotted line. This representation will be retained in subsequent figures, where the thickness variation of the anode-related unit will not be shown again; in fact, this thickness variation is simply the sum of the thickness variations of the silicon-related and graphite-related units.

[0032] As previously stated, in box 1, the cell thickness variation x with respect to the state of charge (SOC) is measured in dynamic in-situ measurement 1 within the vehicle. This is in Figure 3 The diagram of a is similar to Figure 2 The diagrams in the text are again presented separately as charts. Figure 3 The state in graph a here represents the state of cell thickness change at the start of a single cell's lifespan. Next to it... Figure 3 Figure b shows essentially the same curve, namely the change in cell thickness x of the entire cell at a later time point as the individual cell ages more rapidly. The corresponding curve here indicates the unknown superposition of portions of the individual active material under the corresponding aging states (a) and (b).

[0033] Here, the curve also shows some characterizing elements, such as the final charging plateau labeled 4 and the charging plateau labeled 5, and the maximum value labeled 6. These serve as characteristic points or characteristic regions of the curve, which can be seen from... Figure 3 This can be seen in a (at the start of a single battery cell's lifespan), and can also be seen from... Figure 3 This can be seen in b (i.e., when a single battery cell has aged).

[0034] The final difference between charging platform 4 and charging platform 5 is Figure 3In a, it is represented as Δx P,Z.1 For the cell thickness variation x of an aged individual battery cell, the difference between charging platform 5 and the final charging platform 4 is also plotted and labeled as Δx. P,Z,2 Platform lag means that the distance between the final charging platform 4 and the charging platform 5 in the direction of unit thickness variation has been significantly reduced.

[0035] exist Figure 4 a and Figure 4 In diagram b, the curves for the silicon-related aging process are now plotted accordingly. Figure 4 a represents the start of the life cycle of a single battery cell. Figure 4 b represents an aged individual cell. That is, silicon-related aging itself (usually denoted as...) It can lag Δx from the current platform P，Z，2 The platform lag Δx with the new individual battery cell P,Z，1 The ratio is determined by multiplying this value by 100 to obtain the percentage value of silicon aging or the degree of silicon aging. Now, in Figure 4 As can be seen in a and b, the plateau hysteresis Δx is determined based on the cell thickness variation of a single battery cell. P,Z，1 Δx P,Z，2 The value corresponds to the silicon curve Δx P，Si，1 Δx P，Si，2 The value in. This means:

[0036]

[0037] This correspondence now makes it possible to compress in the direction of the element thickness variation, from Figure 2 The entire curve is derived from the silicon portion with known unit thickness variations.

[0038] Similar to the previous illustration, Figure 5 a and 5b now consider graphite-related aging. The graphite-related aging, or graphite-related cell thickness variation x, is characterized by the level of the final charging plateau 4 at the zero baseline of cell thickness variation. That is, the solid line represents the height x of the final charging plateau 4 of the overall curve. P,Ent,2 The graphite-related height x of the final charging platform 4 P,Gr，2 Silicon-related height x of the final charging platform 4 P,Gr,1 It is formed by adding them together.

[0039]

[0040] Theoretically speaking, the thickness variation x of the NMC-related unit in the cathode is... P,NMC，2It also plays a very minor role. However, in practice, it is so small that it can be completely ignored in the area of ​​the final charging platform 4.

[0041] That is, in Figure 4 Given the silicon-related height of discharge platform 4 within the context described within the framework, the graphite-related platform height x can now be determined mathematically. P,Gr,2 .

[0042] x P,Gr,2 =x P,Ent,2 -x P,Si,2

[0043] Then, graphite aging. Able to rise again from the platform height of the aging cell x P,Gr,2 Platform height at the start of battery life x P,Gr,1 The ratio is derived from the percentage value, which is then multiplied by 100.

[0044]

[0045] That is, similar to the previous aging curves for individual silicon cell aging states, it is now possible to compress graphite based on the values ​​detected at the final charging platform. Figure 2 Since the initial curves are known in principle, in addition to the curves showing the cell thickness changes of individual cell units, the cell thickness changes related to silicon or graphite can also be recorded simultaneously, which are shown here again in the form of dashed and dotted lines.

[0046] from Figure 2 The diagram also shows the cell thickness variation x used to determine NMC-related changes and thus NMC aging. The characteristic site is the minimum value within a 100% SOC region.

[0047] The known maximum values ​​of silicon-related cell thickness variations and primarily graphite-related cell thickness variations can be offset here. That is, Figure 6 The total maximum value x in b M，Z，2 From the sum

[0048] x M,Z,2 =x M,Si,2 +x M,Gr,2 +x M,NMC,2

[0049] Therefore, x can ultimately be calculated by determining the difference between the maximum value of the entire cell and the two maximum values ​​associated with silicon or graphite. M,NMC,2 .

[0050] x M，NMC,2 =x M，Z，2 -xM，Si,2 -x M，Gr，2

[0051] For NMC-related aging, the following applies:

[0052]

[0053] The same situation exists here: as the minimum value decreases, the aging of the NMC or cathode increases.

Claims

1. A method for determining the aging of an individual battery cell in a lithium-ion battery composed of multiple battery cells, wherein the cell thickness change (x) of the individual battery cell with respect to its state of charge (SOC) is detected to draw conclusions about the aging of the individual battery cell. Its features are, The curve of the cell thickness variation (x) is evaluated over the entire state of charge (SOC) range during the charging and discharging of the respective individual battery cell, such that the characterizing properties of the curve are used to infer the degree of participation of at least one of the active materials (Si, Gr, NMC) in the cell thickness variation (x) and thus infer the degree of participation in the aging.

2. The method according to claim 1, Its features are, The aging of at least silicon (Si), graphite (Gr), and NMC is deduced, with the premise that there is a linear relationship between the volume fraction of the corresponding active materials (Si, Gr, NMC) and their effect on the change in the unit thickness (x).

3. The method according to claim 1 or 2, Its features are, To determine the silicon aging (θ) Si The platform hysteresis (Δx) of the unit thickness change (x) between the charging platform (5) and the final charging platform (4) is evaluated. P ).

4. The method according to claim 1, 2 or 3, Its features are, To determine the graphite aging (θ) Gr The height (x) of the final charging platform (4) is evaluated based on the change in the unit thickness (x). P ).

5. The method according to any one of claims 1 to 4, Its features are, To determine the NMC aging (θ) NMC ), evaluate the maximum value (x) of the cell thickness variation (x) with respect to the state of charge (SOC). M ).

6. The method according to claims 3, 4 and 5, Its features are, The silicon aging (θ) is first evaluated using the portion of the detected change in cell thickness (x). Si Then, the aging of the graphite (θ) was evaluated. Gr Finally, the NMC aging (θN) was evaluated. MC ).

7. The method according to any one of claims 3 to 6, Its features are, Using the hysteresis (Δx) from the platform P At least one value determined in ) is used to compress a known initial curve of the cell thickness variation (x) related to the silicon state of charge (SOC) in the direction of the cell thickness variation (x) in order to obtain a curve of the cell thickness variation (x) corresponding to the current aging over the entire SOC range during charging and discharging.

8. The method according to any one of claims 4 to 7, Its features are, Using the height (x) from the discharge platform (4) P At least one value determined in ) is used to compress a known initial curve of the cell thickness variation (x) related to the state of charge (SOC) of graphite in the direction of the cell thickness variation (x) in order to obtain a curve of the cell thickness variation (x) corresponding to the current aging over the entire SOC range during charging and discharging.

9. The method according to any one of claims 3 to 8, Its features are, To determine the silicon aging (θ) si The unit thickness variation (x) is evaluated in relation to the platform hysteresis (Δx) between the charging platform (5) and the final charging platform (4). P ), where the platform lags behind (Δx) P When the θ decreases, the silicon aging is inferred. si This will exacerbate the situation.

10. The method according to claim 9, Its features are, To determine the graphite aging (θ) Gr The height detected from the curve of the final charging platform (4) across the entire unit and the previously determined silicon-related height (x) P,Si The graphite-related height (x) of the final charging platform (4) is determined from the difference between the two values. P,Gr ), wherein the height (x) associated with the reduced graphite is utilized P,Gr ), thus inferring the aging of the graphite (θ) Gr This will exacerbate the situation.

11. The method according to claims 9 and 10, Its features are, In the silicon aging and the graphite aging (θ) si θ Gr Given that the changes in cell thickness (x) caused by silicon (Si) and graphite (Gr) are known, determine the corresponding maximum value (xi) of the cell thickness variation (x) caused by silicon (Si) and graphite (Gr) under the fully charged state. M ), from the total maximum value related to graphite and silicon, the maximum value (x M The minimum value (x) of the cell thickness change caused by NMX in the fully charged state is determined from the difference. M,NMC ), where the minimum value (x) is used M,NMC The decrease in the value of ) is used to infer the aging of NMC (θ). NMC This will exacerbate the situation.

Images