A method, product, and electronic device for predicting battery cycle life

Abstract

This invention discloses a method, product, and electronic device for predicting battery cycle life, belonging to the field of new energy battery technology. The method includes: selecting multiple battery cells for cycle life testing and recording the capacity of each cell; selecting the cell with the lowest capacity as a reference cell, using the number of cycles as the reference number or cycle time as the reference time, and capacity change as the reference value; selecting other cells as test cells, obtaining the battery capacity change data of the test cells in the first j cycles, and calculating the difference in the number of cycles or cycle time between the test cells and the reference cells; calculating the functional relationship between the difference in the number of cycles or cycle time of the test cells and the reference cells; deriving the Cycle_diff relationship for the test cells within the reference SOH range; calculating the SOH of the test cells, and obtaining the number of cycles or cycle time of the test cells under different SOH values. This invention can predict battery cycle life using only a small amount of test data.

Description

Technical Field

[0001] This invention relates to the field of new energy battery technology, and in particular to a method, product and electronic device for predicting battery cycle life. Background Art

[0002] With the rapid development of the new energy industry and the promotion of my country's energy storage policies, energy storage power stations, represented by lithium-ion batteries, have experienced explosive growth in installations in recent years, with increasingly larger installed capacities. At the same time, this has brought about a rise in battery management issues. Among these, lithium-ion batteries experience lifespan degradation and capacity reduction during use due to factors such as battery material decomposition, electrolyte drying, and lithium-ion loss, failing to meet the initial design specifications of the power station. Therefore, estimating the lifespan of lithium batteries in energy storage power stations is a crucial aspect of power station management.

[0003] Current challenges in estimating the lifespan of lithium batteries include: firstly, the need for extensive cell testing across numerous application conditions, resulting in long testing cycles and cumbersome procedures, hindering flexible application; secondly, the use of AI technology to build neural networks, requiring extensive training with high-quality test data, places high demands on the data. Both of these factors limit the development and application of battery lifespan SOH algorithms.

[0004] Therefore, the present invention aims to propose a method, product and electronic device for predicting battery cycle life, which can achieve the prediction of battery cycle life using a small amount of test data. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by providing a method, product, and electronic device for predicting battery cycle life.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0007] In a first aspect, the present invention provides a method for predicting battery cycle life.

[0008] A method for predicting battery cycle life includes the following steps:

[0009] Step 1: Select multiple cells of the same type for cycle life testing. Record the SOH of each cell during the cycle life test. Each cell corresponds to one cycle life test condition.

[0010] Step 2: After the cycle life test, select the cell with the lowest SOH as the reference cell, and use the number of cycles of the reference cell as the reference number. Alternatively, the cycle time of a reference cell can be used as the reference time. Capacity change as ;

[0011] Step 3: Select other cells as cells to be tested, and obtain the battery capacity change data of the cells to be tested in the first j cycles. Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same capacity.

[0012] Step 4: Calculate the cycle count difference (Cycle_diff) for each cell under test and compare it with that of the reference cell. The functional relationship, or the time difference Time_diff between the reference cell and the time difference in the cycle. The functional relationship;

[0013] Step 5: Based on the above functional relationship, the range of the battery cell under test within the reference SOH is determined to be... The corresponding Cycle_diff or Time_diff formula;

[0014] Step 6: Based on the above relationship, calculate the SOH of the cell under test, and obtain the number of cycles or cycle time of the cell under test under different SOHs based on the SOH of the cell under test.

[0015] Further, in step 3, selecting other battery cells as the cells to be tested, obtaining the battery capacity change data of the cells to be tested in the first j cycles, and calculating the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same capacity, specifically includes the following steps:

[0016] Step 31: Establish a functional relationship between the SOH of the cell under test and the number of cycles or cycle time, with SOH as the independent variable and the number of cycles as the inverse variable. Or cycle time As the dependent variable, the functional relationship is: or ;

[0017] Step 32: After establishing the functional relationship, Substituting this into the calculation as the independent variable, we obtain the value of the battery cell under test. The corresponding number of cycles Or cycle time ,Right now or ;

[0018] Step 33: The difference in the number of cycles between each cell under test and the reference cell at the same capacity is Cycle_diff or the difference in cycle time is Time_diff, i.e.:

[0019] or .

[0020] Furthermore, in step 4, the calculation of the cycle difference (Cycle_diff) for each cell under test is compared with that of the reference cell. The functional relationship or cycle time difference Time_diff between the reference cell The functional relationship can be in polynomial, exponential, or power form.

[0021] Further, in step 4, the calculation of the cycle count difference (Cycle_diff) or cycle time difference (Time_diff) for each cell under test is compared with that of the reference cell. The function relationship requires calculation of the parameters to be identified in the function, and the number of values ​​contained in the Cycle_diff or Time_diff array should be greater than or equal to the parameters to be identified in the function relationship.

[0022] Further, in step 6, the SOH of the cell under test is calculated, i.e. The number of cycles corresponding to time The sum of Cycle_diff, or The sum of the time corresponding to the given time and Time_diff.

[0023] Furthermore, in step 5, the cell under test is within the reference SOH range of... The corresponding Cycle_diff or Time_diff relationship is:

[0024] ;

[0025] .

[0026] Furthermore, the cycle life test conditions include the current rate and temperature during the charging and discharging process of the battery cell under test.

[0027] Secondly, the present invention provides a computer program product.

[0028] A computer program product includes a computer program that, when executed by a processor, implements the battery cycle life prediction method described above.

[0029] Thirdly, the present invention provides an electronic device.

[0030] An electronic device includes at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the battery cycle life prediction method described above.

[0031] In summary, compared with the prior art, the beneficial effects of the above technical solution are:

[0032] This invention enables the prediction of battery cycle life using only a small amount of test data. Specifically, it allows the prediction of the lifespan of other cells based on the lifespan test data of one cell, requiring less data, achieving high accuracy, and reducing the complexity of cell lifespan prediction. Attached Figure Description

[0033] Figure 1 Flowchart for predicting battery cycle life;

[0034] Figure 2 A cycle life curve for a set of battery cells;

[0035] Figure 3 The test curves show the cycle life of two battery cells;

[0036] Figure 4 The difference in cycle life between the two cells under different state of equilibrium (SOH) conditions;

[0037] Figure 5 Fitting curves for the difference in cycle life between the two cells;

[0038] Figure 6 This is to predict the cycle life of the battery cell. Detailed Implementation

[0039] The principles and features of the present invention are described below with reference to all the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0040] This invention discloses a battery cycle life prediction method, product, and electronic device.

[0041] In a first aspect, embodiments of the present invention disclose a method for predicting battery cycle life.

[0042] Reference Figures 1-6 A method for predicting battery cycle life includes the following steps:

[0043] Step 1: Select multiple cells of the same type for cycle life testing. Record the capacity change of each cell during the cycle life test. Each cell corresponds to one cycle life test condition.

[0044] Step 2: After the cycle life test, select the cell with the lowest SOH as the reference cell, and use the number of cycles of this cell as the reference number of cycles. Or use the cycle time as a reference time Capacity change as ;

[0045] Step 3: Select other cells as cells to be tested, and obtain the battery capacity change data of the cells to be tested in the first j cycles. Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same capacity.

[0046] Step 4: Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell under test and the reference cell. The functional relationship;

[0047] Step 5: Based on the above functional relationship, the range of the battery cell under test within the reference SOH is determined to be... The corresponding relationship between Cycle_diff or Time_diff is, i.e. ,

[0048] or ;

[0049] Step 6: Based on the above relationship, calculate the SOH of the cell under test, and obtain the number of cycles or cycle time of the cell under test under different SOHs based on the SOH of the cell under test.

[0050] This invention discloses a battery cycle life prediction method that can predict battery cycle life using only a small amount of test data. In this embodiment, the lifespan of other cells can be predicted from the lifespan test data of one cell, requiring less data, achieving high accuracy, and reducing the complexity of cell lifespan prediction.

[0051] The following is a detailed explanation of each of the above steps.

[0052] Step 1: Select multiple cells of the same type for cycle life testing. Record the capacity change of each cell during the cycle life test. Each cell corresponds to one cycle life test condition. The cycle life test conditions include the current rate and temperature during the charging and discharging process of the cell under test.

[0053] Specifically, multiple cells of the same type were selected for life testing. Each cell corresponded to a specific cycle life test condition. During the test, the capacity change of each cell with cycle time or number of cycles was recorded, as shown in Table 1. (Number of cycles for each cell is also listed.) or cycle time They can be different, but the corresponding SOH values ​​must be less than the set threshold. Threshold The setting needs to take into account the SOH degradation of the battery cells. When each battery cell exhibits significant degradation and the SOH varies considerably between cells, this invention will achieve better results. In this invention, the threshold... It can be set to 0.93, meaning that the SOH of all cells must be less than 0.93.

[0054] Table 1. Test capacity variation data for each cell

[0055]

[0056] Step 2: After the cycle life test, select the cell with the lowest SOH as the reference cell, and use the number of cycles of this cell as the reference number of cycles. Or use the cycle time as a reference time Capacity change as .

[0057] Specifically, Figure 2 This is a set of cycle life test curves for battery cells. Each cell corresponds to different cycle life test conditions, thus exhibiting different capacity degradation trajectories. Since the operating conditions of batteries in actual use are more complex, their capacity degradation curves will be even more complex. The number of cycles in each cycle life test and the corresponding State of Harm (SOH) value are recorded. From this set of cells, the cell with the lowest SOH is selected as the reference cell, and another cell is selected as the cell to be tested.

[0058] The reference cell is the one with the lowest SOH (State of Health), not any arbitrary cell. For example, cell 1 has an SOH range of [1~0.83], and cell 2 has an SOH range of [1~0.84]. It can be seen that cell 2 has no cycle count or cycle time within the SOH range of [0.84~0.83]. This invention uses the SOH variation data of cell 1 to predict the cycle count or cycle time corresponding to an SOH of [0.84-0.83] for cell 2. Therefore, the cell with the lowest SOH must be selected.

[0059] Step 3: Select other cells as the cells to be tested, and obtain the battery capacity change data of the cells to be tested in the first j cycles. Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same capacity.

[0060] Specifically, in step 2, after excluding the reference cell, N-1 other cells are selected as the cells to be tested, and the battery capacity change data of the first j (j < i) cycles of the cells to be tested are extracted. The test curves of the two selected cells are shown in the figure. Figure 3 As shown, it can be seen that, under the same SOH, the reference cell has more cycle times than the cell under test.

[0061] Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell under test and the reference cell at the same capacity, as shown in Tables 2 and 3.

[0062] Table 2. Difference in cycle count between each cell and the reference cell at the same capacity.

[0063]

[0064] Table 3. Time difference between each battery cell and the reference battery cell at the same capacity.

[0065]

[0066] Because in each of the cells under test There may not be a corresponding number of loops or loop time for each loop; in this case, it is necessary to calculate the number of loops or loop time for each loop. At that time, the number of cycles or cycle time corresponding to the battery cell under test. The calculation process is as follows:

[0067] Step 31: Establish a functional relationship between the SOH of the cell under test and the number of cycles, with SOH as the independent variable and the number of cycles as the inverse variable. Or cycle time As the dependent variable, the functional relationship is: or ;

[0068] Step 32: After establishing the functional relationship, Substituting this into the calculation as the independent variable, we obtain the value of the battery cell under test. The corresponding number of cycles Or cycle time ,Right now or ;

[0069] For example, a battery undergoing one full charge and then full discharge constitutes one charge-discharge cycle, or one cycle. If another full charge and full discharge cycle is performed, it becomes two cycles. The cycle time is the duration the battery operates. If a full charge and full discharge cycle takes two hours, then the cycle time is two hours.

[0070] Step 33: The difference in the number of cycles between each cell under test and the reference cell at the same capacity is Cycle_diff or the difference in cycle time is Time_diff, i.e.:

[0071] or .

[0072] Step 4: Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell under test and the reference cell. The functional relationship.

[0073] Specifically, test data of the reference cell and the cell under test were selected with a SOH range of [0.93~1]. The difference in the number of cycles between the cell under test and the reference cell at the same capacity was calculated as Cycle_diff. The calculation results are as follows: Figure 4 As shown in the figure, the difference in the number of cycles between the two cells gradually stabilizes as the state of oxygen (SOH) decreases.

[0074] according to Figure 4 The results are used to calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between the tested cell and the reference cell. The functional relationship, that is or The parameters to be identified in the function need to be calculated, and the number of values ​​contained in the Cycle_diff or Time_diff array should be greater than or equal to the parameters to be identified in the function expression.

[0075] The functional relationship in this embodiment is as follows: a, b, and c are the parameters to be identified. After data fitting, a = 244.38, b = 38.78, and c = -225.09 were obtained. The fitting results are as follows: Figure 5 As shown, the fitted curve can accurately capture the changes in the original data, demonstrating a good fitting effect. Thus, we have established the mathematical relationship between the cycle difference between the reference cell and the cell under test and the State of Hysteresis (SOH).

[0076] Step 5: Based on the above functional relationship, the range of the battery cell under test within the reference SOH is determined to be... The corresponding Cycle_diff relation, i.e. .

[0077] Specifically, since the cycle life test result of the reference cell was 0.83, while the lowest SOH value used in the aforementioned fitting curve was 0.93, in order to predict the number of cycles of the cell under test within the SOH range of [0.83~0.93], the SOH value of the reference cell was substituted into the formula, resulting in the following equation:

[0078] .

[0079] Step 6: Based on the above relationship, calculate the SOH of the cell under test, and obtain the number of cycles or cycle time of the cell under test under different SOHs based on the SOH of the cell under test.

[0080] Specifically, the SOH of the cell to be predicted is calculated, i.e. The number of cycles corresponding to time The sum of Cycle_diff, or The number of cycles corresponding to time The sum of Time_diff is calculated using the following formula: ;

[0081] or .

[0082] This allows us to obtain the number of cycles of the battery cell under different SOH conditions, such as... Figure 6 As shown, from Figure 6 As can be seen, the predicted lifespan of the tested battery cell almost matches the actual value, indicating a good prediction effect.

[0083] Secondly, embodiments of the present invention also provide a computer program product.

[0084] A computer program product includes a computer program that, when executed by a processor, implements the battery cycle life prediction method described above.

[0085] Thirdly, embodiments of the present invention also provide an electronic device.

[0086] An electronic device includes at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the battery cycle life prediction method described above.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting battery cycle life, characterized in that, Includes the following steps: Step 1: Select multiple cells of the same type for cycle life testing. Record the SOH of each cell during the cycle life test. Each cell corresponds to one cycle life test condition. Step 2: After the cycle life test, select the cell with the lowest SOH as the reference cell, and use the number of cycles of the reference cell as the reference number. Alternatively, the cycle time of a reference cell can be used as the reference time. SOH change as ; Step 3: Select other cells as the cells to be tested, and obtain the battery SOH change data of the cells to be tested in the first j cycles. Calculate the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same SOH. Step 4: Calculate the cycle count difference (Cycle_diff) for each cell under test and compare it with that of the reference cell. The functional relationship, or the time difference Time_diff between the reference cell and the time difference in the cycle. The functional relationship; Step 5: Based on the above functional relationship, the range of the battery cell under test within the reference SOH is determined to be... The corresponding Cycle_diff or Time_diff formula; Step 6: Based on the above relationship, determine the number of cycles or cycle time of the battery cell under different SOH conditions; In step 3, selecting other battery cells as the cells to be tested and obtaining the battery state of harmonics (SOH) change data of the first j cycles of the cells to be tested, and calculating the difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell to be tested and the reference cell at the same SOH, specifically includes the following steps: Step 31: Establish a functional relationship between the SOH of the cell under test and the number of cycles or cycle time, with SOH as the independent variable and the number of cycles as the inverse variable. Or cycle time As the dependent variable, the functional relationship is: or ; Step 32: After establishing the functional relationship, Substituting this as the independent variable into the calculation, we obtain the value of the battery cell under test. The corresponding number of cycles Or cycle time ,Right now or ; Step 33: The difference in the number of cycles (Cycle_diff) or the difference in cycle time (Time_diff) between each cell under test and the reference cell at the same state of equilibrium (SOH), i.e.: or ; The calculation of the cycle count difference (Cycle_diff) or cycle time difference (Time_diff) for each tested cell is compared with that of the reference cell. The function relationship requires calculation of the parameters to be identified in the function, and the number of values ​​contained in the Cycle_diff or Time_diff array should be greater than or equal to the parameters to be identified in the function relationship. In step 6, the formula for calculating the number of cycles or cycle time of the battery cell under different SOH conditions is as follows: ； or ; In step 5, the cell under test is within the reference SOH range of The corresponding Cycle_diff or Time_diff relationship is: 。 2. The battery cycle life prediction method according to claim 1, characterized in that: In step 4, the difference in the number of cycles (Cycle_diff) for each cell under test is calculated and compared with that of the reference cell. The functional relationship or cycle time difference Time_diff between the reference cell The functional relationship can be in polynomial, exponential, or power form.

3. The battery cycle life prediction method according to claim 1, characterized in that: Cyclic life test conditions include current rate and temperature during the charging and discharging process of the battery cell under test.

4. A computer program product, characterized in that: The method includes a computer program that, when executed by a processor, implements a battery cycle life prediction method according to any one of claims 1-3.

5. An electronic device, characterized in that: The device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a battery cycle life prediction method according to any one of claims 1-3.

Images