Battery cell sorting method and device, electronic equipment and computer readable storage medium

Abstract

The invention discloses a battery cell sorting method and device, electronic equipment and a computer readable storage medium, and belongs to the field of batteries. The method comprises the following steps: acquiring index data of a first battery cell; for each battery cell set in a first preset number of battery cell sets, determining a first distance between the first battery cell and a clustering center based on the index data of the first battery cell and the index data of the clustering center of the battery cell set, the first preset number of battery cell sets are determined by clustering in advance based on a first training data set, and the first training data set comprises the index data of a second preset number of sample battery cells; the first battery cells are sorted into a target battery cell set, and the target battery cell set is the battery cell set with the minimum first distance in the first preset number of battery cell sets. By adopting the scheme of the invention, the production efficiency of the battery can be improved.

Description

Technical Field

[0001] This application relates to the field of batteries, and more particularly to a cell sorting method, apparatus, electronic device, and computer-readable storage medium. Background Technology

[0002] Maintaining battery consistency is crucial for ensuring the performance, lifespan, and safety of battery packs. Cell sorting is a vital step in the battery packing process to maintain battery consistency. Specifically, it involves grouping cells with similar properties into groups, and then assembling cells from the same group into a single battery, thereby controlling battery consistency.

[0003] In related technologies, it is usually necessary to obtain the relevant indicators of all cells to be sorted, and then to perform clustering and sorting of all cells to be sorted in a unified manner. After the sorting is completed, battery production can proceed. However, this sorting method will affect the battery production efficiency. Summary of the Invention

[0004] This application discloses a cell sorting method, apparatus, electronic device, and computer-readable storage medium, which can improve battery production efficiency.

[0005] To solve the above problems, this application adopts the following technical solution: In a first aspect, embodiments of this application disclose a battery cell sorting method, comprising: acquiring indicator data of a first battery cell; for each of a first preset number of battery cell sets, determining a first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set, wherein the first preset number of battery cell sets are pre-determined based on clustering of a first training dataset, the first training dataset including the indicator data of a second preset number of sample battery cells; sorting the first battery cell to a target battery cell set, wherein the target battery cell set is the battery cell set with the smallest first distance among the first preset number of battery cell sets.

[0006] Optionally, the indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time. Determining the first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the cluster center of the battery cell set includes: determining the open-circuit voltage distance between the first battery cell and the cluster center based on the open-circuit voltage of the first battery cell and the open-circuit voltage of the cluster center of the battery cell set; determining the K-value distance between the first battery cell and the cluster center based on the K-value of the first battery cell and the K-value of the cluster center; determining the AC internal resistance distance between the first battery cell and the cluster center based on the AC internal resistance of the first battery cell and the AC internal resistance of the cluster center; determining the storage time distance between the first battery cell and the cluster center based on the storage time of the first battery cell and the storage time of the cluster center; and determining the first distance between the first battery cell and the cluster center based on the open-circuit voltage distance, the K-value distance, the AC internal resistance distance, and the storage time distance.

[0007] Optionally, after sorting the first cell into the target cell set, the method further includes: updating the cluster center of the target cell set based on the updated index data of each cell in the target cell set; obtaining the index data of the second cell; using the second cell as the first cell, and returning to execute the step of determining a first distance between the first cell and the cluster center for each of the first preset number of cell sets, based on the index data of the first cell and the index data of the cluster center of the cell set.

[0008] Optionally, sorting the first battery cell into the target battery cell set includes: for each of the last three preset number of target battery cells added to the target battery cell set, determining a second distance between the first battery cell and the target battery cell based on the indicator data of the first battery cell and the indicator data of the target battery cell, wherein the third preset number is greater than the number of battery cells required for the battery; and sorting the first battery cell into the target battery cell set if none of the third preset number of second distances is greater than a preset threshold.

[0009] Optionally, the indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time; the second distance includes open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance; the step of sorting the first cell to the target cell set when the third preset number of the second distances are all not greater than a preset threshold includes: sorting the first cell to the target cell set when the third preset number of the open-circuit voltage distances are all not greater than a first threshold, the third preset number of the K-value distances are all not greater than a second threshold, the third preset number of the AC internal resistance distances are all not greater than a third threshold, and the third preset number of the storage time distances are all not greater than a fourth threshold.

[0010] Optionally, before obtaining the indicator data of the first battery cell, the method further includes: obtaining the first training dataset; performing clustering based on the indicator data of the sample battery cells in the first training dataset to determine the first preset number of initial battery cell sets; filtering out sample battery cells from the target initial battery cell set from the first training dataset to determine the second training dataset, wherein the target initial battery cell set is the initial battery cell set in the first preset number of initial battery cell sets where the number of battery cells is less than a fifth threshold; performing iterative clustering based on the second training dataset until a stopping iteration condition is met to determine the first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, wherein the preset condition includes the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0011] Optionally, before obtaining the indicator data of the first battery cell, the method further includes: obtaining the first training dataset; for each sample battery cell in the first training dataset, determining the average distance between the sample battery cell and the other sample battery cells based on the indicator data of the sample battery cell and the indicator data of other sample battery cells in the first training dataset; filtering out target sample battery cells from the first training dataset to determine a second training dataset, wherein the target sample battery cells are sample battery cells whose average distance is greater than a seventh threshold; performing iterative clustering based on the second training dataset until a stopping iteration condition is met to determine a first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, the preset condition including the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0012] Optionally, before performing iterative clustering based on the second training dataset, the method further includes: normalizing the indicator data of the sample cells in the second training dataset based on the maximum and minimum values ​​of the indicator data of the sample cells in the second training dataset.

[0013] Secondly, embodiments of this application disclose a battery cell sorting device, comprising: an acquisition module for acquiring indicator data of a first battery cell; a determination module for determining, for each of a first preset number of battery cell sets, a first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set, wherein the first preset number of battery cell sets are pre-determined based on a first training dataset, the first training dataset including the indicator data of a second preset number of sample battery cells; and a sorting module for sorting the first battery cell to a target battery cell set, wherein the target battery cell set is the battery cell set with the smallest first distance among the first preset number of battery cell sets.

[0014] Thirdly, embodiments of this application provide an electronic device including a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer-executable program or instructions, which, when executed by a computer, implement the steps of the method described in the first aspect.

[0016] Fifthly, embodiments of this application provide a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the steps of the method described in the first aspect.

[0017] The technical solution adopted in this application can achieve the following beneficial effects: This application provides a battery cell sorting method. By acquiring the indicator data of a first battery cell, and for each of the first preset number of battery cell sets, based on the indicator data of the first battery cell and the indicator data of the cluster centers of the battery cell sets, a first distance is determined between the first battery cell and the cluster centers. The first preset number of battery cell sets are pre-determined based on clustering of a first training dataset, which includes the indicator data of a second preset number of sample battery cells. Then, the first battery cell is sorted into the target battery cell set with the smallest first distance among the first preset number of battery cell sets. Using this method, the first battery cell can be grouped immediately after acquiring its indicator data, eliminating the need to acquire the indicator data of all cells to be sorted before clustering and sorting. Compared to related technologies that require acquiring the relevant indicators of all cells to be sorted before uniformly clustering and sorting them, this method improves battery production efficiency. Attached Figure Description

[0018] Figure 1 This is a schematic flowchart of a battery cell sorting method disclosed in an embodiment of this application; Figure 2 This is a flowchart of a battery cell sorting process disclosed in an embodiment of this application; Figure 3 This is another flowchart of cell sorting disclosed in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a battery cell sorting device disclosed in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0020] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the electrically connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0021] The following description, in conjunction with the accompanying drawings, details the battery cell sorting method, apparatus, electronic device, and computer-readable storage medium disclosed in this application through specific embodiments and application scenarios.

[0022] This application discloses a method for sorting battery cells. Figure 1 This is a schematic flowchart of a battery cell sorting method disclosed in an embodiment of this application. Figure 1 As shown, the method includes the following steps: S120: Obtain the performance data of the first battery cell.

[0023] This indicator data may include open-circuit voltage, K value, AC internal resistance, and storage time.

[0024] In this application, the first cell is the cell currently to be sorted.

[0025] The open circuit voltage (OCV) of a battery cell refers to the potential difference between the positive and negative terminals when the battery is not connected to any load (i.e., no current flows through it) and has been left to stand for a sufficient period of time to reach a stable state.

[0026] The K-value of a battery cell is a core indicator for measuring the self-discharge rate of a lithium-ion battery. It is defined as the decrease in open-circuit voltage (OCV) per unit time, usually measured in mV / d (millivolts per day) or mV / h (millivolts per hour). It directly reflects the internal chemical stability and reliability of the battery.

[0027] The alternating current internal resistance (ACIR) of a battery cell is a core parameter that measures the battery's ability to impede current under alternating current excitation. It reflects the comprehensive impedance characteristics inside the battery and is a key indicator for evaluating the battery's dynamic performance, health status, and safety.

[0028] S140. For each of the first preset number of battery cell sets, based on the index data of the first battery cell and the index data of the cluster center of the battery cell set, determine the first distance between the first battery cell and the cluster center.

[0029] The first preset number of battery cells is determined in advance based on clustering of a first training dataset, and the first training dataset includes the indicator data of a second preset number of sample battery cells.

[0030] It should be noted that the specific values ​​of the first and second preset quantities can be set according to actual needs, and this application does not impose specific limitations on them. A first training dataset is constructed based on the indicator data of the second preset quantity of sample cells.

[0031] In this application, iterative clustering is performed in advance based on the index data of a second preset number of sample cells to determine a first preset number of cell sets. Then, with the index data of the first cell obtained, the first distance between the cluster centers of the first cell and each cell set in the first preset number of cell sets is calculated.

[0032] It should be noted that the first distance between the first cell and the cluster center in this application can be an Euclidean distance.

[0033] S160. The first battery cell is sorted into the target battery cell set.

[0034] The target cell set is the cell set with the smallest first distance among the first preset number of cell sets.

[0035] In other words, the first battery cell is sorted into the target battery cell set whose cluster center is closest to the first battery cell in the first preset number of battery cell sets.

[0036] By adopting the scheme of this application, a first preset number of cell sets can be obtained by clustering the index data of a small number of sample cells. Then, according to the production process, when the index data of the first cell to be sorted is obtained, the first distance between the first cell and the cluster center of each cell set in the first preset number of cell sets is calculated. The first cell is then sorted into the target cell set in the first preset number of cell sets whose cluster center is closest to the first cell. Since this application can group the first cell after obtaining the index data of the first cell, compared with the related technology, which requires obtaining the relevant indexes of all cells to be sorted and then uniformly clustering and sorting all cells to be sorted, the production efficiency of batteries can be improved.

[0037] This application provides a battery cell sorting method. By acquiring the indicator data of a first battery cell, and for each of the first preset number of battery cell sets, based on the indicator data of the first battery cell and the indicator data of the cluster centers of the battery cell sets, a first distance is determined between the first battery cell and the cluster centers. The first preset number of battery cell sets are pre-determined based on clustering of a first training dataset, which includes the indicator data of a second preset number of sample battery cells. Then, the first battery cell is sorted into the target battery cell set with the smallest first distance among the first preset number of battery cell sets. Using this method, the first battery cell can be grouped immediately after acquiring its indicator data, eliminating the need to acquire the indicator data of all cells to be sorted before clustering and sorting. Compared to related technologies that require acquiring the relevant indicators of all cells to be sorted before uniformly clustering and sorting them, this method improves battery production efficiency.

[0038] In one implementation, the indicator data may include open-circuit voltage, K-value, AC internal resistance, and storage time. Determining the first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set may include: determining an open-circuit voltage distance between the first battery cell and the cluster center based on the open-circuit voltage of the first battery cell and the open-circuit voltage of the cluster center of the battery cell set; determining a K-value distance between the first battery cell and the cluster center based on the K-value of the first battery cell and the K-value of the cluster center; determining an AC internal resistance distance between the first battery cell and the cluster center based on the AC internal resistance of the first battery cell and the AC internal resistance of the cluster center; determining a storage time distance between the first battery cell and the cluster center based on the storage time of the first battery cell and the storage time of the cluster center; and determining the first distance between the first battery cell and the cluster center based on the open-circuit voltage distance, the K-value distance, the AC internal resistance distance, and the storage time distance.

[0039] For example, the first distance between the first cell and the cluster center can be determined by the following formula; ,in, The first distance is between the first battery cell and the cluster center. This is the open-circuit voltage of the first battery cell. This is the open-circuit voltage of the cluster center. The open-circuit voltage distance between the first cell and the cluster center. The K value of the first cell. The K value is the value of the cluster center. The distance between the first battery cell and the cluster center is represented by the K-value distance. The AC internal resistance of the first battery cell. This represents the internal resistance to communication within the cluster center. The distance between the first battery cell and the cluster center is the AC internal resistance distance. The storage time for the first battery cell. The storage time for this cluster center. This represents the storage time distance between the first battery cell and the cluster center.

[0040] Alternatively, after determining the open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance between the first cell and the cluster center, the absolute values ​​of the open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance can be summed to obtain the first distance between the first cell and the cluster center.

[0041] In one implementation, after sorting the first battery cell into the target battery cell set, the method may further include: updating the cluster center of the target battery cell set based on the updated index data of each battery cell in the target battery cell set; obtaining the index data of the second battery cell; using the second battery cell as the first battery cell, and returning to execute the step of determining a first distance between the first battery cell and the cluster center for each of the first preset number of battery cell sets, based on the index data of the first battery cell and the index data of the cluster center of the battery cell set.

[0042] In other words, after sorting the first battery cell into the target battery cell set, the cluster center of the target battery cell set is recalculated and updated. Then, when the index data of the second battery cell is subsequently obtained and the second battery cell needs to be sorted, the first distance between the cluster center of the target battery cell set and the second battery cell is calculated based on the updated cluster center of the target battery cell set. By adopting the scheme of this application, the cluster center of the battery cell set is updated in real time after a new battery cell is added. This can adapt to the offset changes of new data, avoid the cluster center not being representative of the global group center due to too few sample battery cells, avoid inconsistencies caused by the deviation between sampled data and overall data, improve the consistency of battery cell sorting, and enhance robustness.

[0043] In this application, the cluster centers of the cell set are calculated based on the index data of all cells in the cell set. For example, the average open-circuit voltage, average K value, average AC internal resistance, and average storage time of all cells in the cell set are calculated respectively to obtain the cluster centers of the cell set (average open-circuit voltage, average K value, average AC internal resistance, and average storage time). The calculation formula is as follows: ,in, This represents the cluster center of the k-th cell set. This represents the set of the k-th battery cells. This represents the number of battery cells in the k-th battery cell set. This represents the feature vector (which can be multi-dimensional) of the i-th cell.

[0044] It should be noted that this cyclical process continues until all cells in that capacity group have been grouped.

[0045] In one implementation, sorting the first battery cell to the target battery cell set may include: for each of the last three preset number of target battery cells added to the target battery cell set, determining a second distance between the first battery cell and the target battery cell based on the indicator data of the first battery cell and the indicator data of the target battery cell, wherein the third preset number is greater than the number of battery cells required for the battery; and sorting the first battery cell to the target battery cell set if none of the third preset number of second distances is greater than a preset threshold.

[0046] Since this solution can produce batteries while sorting cells, thereby improving battery production efficiency, in order to ensure the quality of the produced batteries, the second distance between the first cell and the third preset number of target cells added to the target cell set is calculated. If the calculated third preset number of second distances are not greater than a preset threshold, the first cell is sorted into the target cell set; otherwise, the first cell is discarded, so as to ensure that the attribute differences between each battery cell are not too large.

[0047] It should be noted that the specific calculation method for the second distance here is similar to that for the first distance mentioned above, and will not be repeated here. The larger the value of the third preset number, the more battery cells are discarded, and the higher the fault tolerance. The specific value of the third preset number can be set according to actual needs, and this application does not impose any specific limitations on it.

[0048] In one implementation, the indicator data may include open-circuit voltage, K-value, AC internal resistance, and storage time; the second distance may include open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance; the step of sorting the first cell to the target cell set when the third preset number of the second distances are all not greater than a preset threshold may include: sorting the first cell to the target cell set when the third preset number of the open-circuit voltage distances are all not greater than a first threshold, the third preset number of the K-value distances are all not greater than a second threshold, the third preset number of the AC internal resistance distances are all not greater than a third threshold, and the third preset number of the storage time distances are all not greater than a fourth threshold.

[0049] In other words, if the distances between the third preset number of target cells added last in the first cell and the target cell set are all no greater than the first threshold, the distances between the third preset number of open-circuit voltages are all no greater than the second threshold, the distances between the third preset number of K values ​​are all no greater than the third threshold, and the distances between the third preset number of AC internal resistances are all no greater than the third threshold, and the distances between the third preset number of storage times are all no greater than the fourth threshold, then the first cell will be sorted into the target cell set; otherwise, the first cell will be discarded to ensure that the attribute differences between each battery cell are not too large.

[0050] It should be noted that this application does not limit the specific values ​​of the first threshold, the second threshold, the third threshold, and the fourth threshold; these values ​​can be set according to actual needs.

[0051] In one implementation, before obtaining the indicator data of the first battery cell, the method may further include: obtaining the first training dataset; performing clustering based on the indicator data of the sample battery cells in the first training dataset to determine a first preset number of initial battery cell sets; filtering out sample battery cells from the target initial battery cell set from the first training dataset to determine a second training dataset, wherein the target initial battery cell set is an initial battery cell set in the first preset number of initial battery cell sets where the number of battery cells is less than a fifth threshold; performing iterative clustering based on the second training dataset until a stopping iteration condition is met to determine the first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, the preset condition including the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0052] In other words, after filtering out abnormal cells from the first training dataset to obtain the second training dataset, this application performs iterative clustering based on the second training dataset to determine a first preset number of cell sets. Clustering is an unsupervised machine learning method whose purpose is to group similar data points into one group and dissimilar data points into different groups.

[0053] This application uses the kmenes++ algorithm to cluster sample cells in a first training dataset, obtaining a first preset number of initial cell sets. Then, it checks the number of cells in each initial cell set, considering sets with fewer than a fifth threshold as abnormal sets. These abnormal sets are removed from the first training dataset, resulting in a second training dataset. The kmenes++ algorithm is then used to select the first preset number of sample cells from the second training dataset as cluster centers. The remaining sample cells in the second training dataset are then assigned to the cell sets belonging to the nearest cluster centers. The cluster centers for each cell set are recalculated, and it is determined whether the iteration stopping condition is met. If not, the iterative clustering process based on the second training dataset continues. If the condition is met, the first preset number of cell sets are determined. The iteration stopping condition in this application can be meeting a preset number of iterations or the cluster centers no longer showing significant changes.

[0054] In this application, the method for determining the initial cluster centers based on the kmeans++ algorithm is as follows: S1. Randomly select a battery cell from the dataset as the first cluster center. S2. For each battery cell in the dataset, calculate its shortest distance to the currently selected center point, and calculate the probability of each battery cell being selected as the next center point based on the square of these distances. S3. Use the roulette wheel method to select the next center point based on probability, that is, data points farther away from the existing center points have a higher probability of being selected as the new center point. Repeat S2 and S3 until a first preset number of cluster centers are selected.

[0055] For example, the fifth threshold can be 5. This application does not specifically limit the value of the fifth threshold, and it can be adjusted according to actual needs.

[0056] In one implementation, before obtaining the indicator data of the first battery cell, the method may further include: obtaining the first training dataset; for each sample battery cell in the first training dataset, determining the average distance between the sample battery cell and the other sample battery cells based on the indicator data of the sample battery cell and the indicator data of other sample battery cells in the first training dataset; filtering out target sample battery cells from the first training dataset to determine a second training dataset, wherein the target sample battery cells are sample battery cells whose average distance is greater than a seventh threshold; performing iterative clustering based on the second training dataset until a stopping iteration condition is met to determine a first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, the preset condition including the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0057] In other words, after filtering out abnormal cells in the first training dataset and obtaining the second training dataset, this application performs iterative clustering based on the second training dataset to determine a first preset number of cell sets.

[0058] This application can calculate the distance between a sample cell and other sample cells in the first training dataset, and then calculate the average distance between the sample cell and other sample cells in the first training dataset. Sample cells with an average distance greater than a seventh threshold are considered abnormal cells and are removed from the first training dataset to obtain a second training dataset. Then, the k-means++ algorithm is used to select a first preset number of sample cells from the second training dataset as cluster centers. The other sample cells in the second training dataset are then assigned to the cell sets belonging to the nearest cluster centers. The cluster centers of each cell set are recalculated, and it is determined whether the iteration stopping condition is met. If not, the iterative clustering process based on the second training dataset continues. If the condition is met, the first preset number of cell sets are determined. The iteration stopping condition in this application can be meeting a preset number of iterations or the cluster centers no longer changing significantly.

[0059] In this application, the method for determining the initial cluster centers based on the kmeans++ algorithm is as follows: S1. Randomly select a battery cell from the dataset as the first cluster center. S2. For each battery cell in the dataset, calculate its shortest distance to the currently selected center point, and calculate the probability of each battery cell being selected as the next center point based on the square of these distances. S3. Use the roulette wheel method to select the next center point based on probability, that is, data points farther away from the existing center points have a higher probability of being selected as the new center point. Repeat S2 and S3 until a first preset number of cluster centers are selected.

[0060] In one implementation, before performing iterative clustering based on the second training dataset, the method may further include: normalizing the indicator data of the sample cells in the second training dataset based on the maximum and minimum values ​​of the indicator data of the sample cells in the second training dataset.

[0061] In other words, the index data (open circuit voltage, K value, AC internal resistance, and storage time) of the sample cells in the second training dataset are first normalized, and then iterative clustering is performed.

[0062] For example, when normalizing the open-circuit voltage of the sample cells in the second training dataset, the open-circuit voltage of the current sample cell can be normalized based on the maximum and minimum open-circuit voltages among all sample cells in the second training dataset. For example, the normalization can be performed using the following formula: ,in, This represents the value obtained after normalization. This indicates the open-circuit voltage of the current sample cell. Indicates the minimum open-circuit voltage. This indicates the maximum open-circuit voltage.

[0063] The K value, AC internal resistance, and storage time normalization methods are similar to those for open-circuit voltage, and will not be repeated here.

[0064] It should be noted that after obtaining the index data of the first and second battery cells, the data can be processed according to the normalization method described above before battery cell sorting.

[0065] like Figure 2 and Figure 3 As shown, the cell sorting method of this application includes the following steps: The Manufacturing Execution System (MES) acquires the indicator data of the cells and sends it to the server. The server determines whether to perform iterative clustering training. If so, it stores the data in the database, filters out abnormal cells, normalizes the indicator data of the cells, and then performs iterative clustering to generate a clustering model. If not, it normalizes the indicator data of the cells, and then determines the cell sorting result based on the clustering model and returns it to the automatic sorting machine. Furthermore, it performs a distance check on the cell and the cells in the corresponding sorted cell set to ensure that the attribute differences between each battery cell are not too large, thereby ensuring the efficiency and quality of the generated battery. Additionally, after completing the cell sorting, the cluster centers of the cell set added to the cell are updated. It should be noted that the clustering model here is the first preset number of cell sets mentioned above.

[0066] Furthermore, this application uses a clustering algorithm for cell sorting, fully considering the cell structure itself, improving consistency while increasing production efficiency. The method in this application can group cells according to the data distribution of batches, improving cell consistency and sorting efficiency.

[0067] The battery cell sorting method provided in this application can be executed by a battery cell sorting device. This application uses a battery cell sorting device executing the battery cell sorting method as an example to illustrate the battery cell sorting device provided in this application.

[0068] Figure 4 This is a schematic diagram of the structure of a battery cell sorting device disclosed in an embodiment of this application. Figure 4 As shown, the battery cell sorting device 400 includes: an acquisition module 410, a determination module 420, and a sorting module 430.

[0069] In this application, the acquisition module 410 is used to acquire the indicator data of the first battery cell; the determination module 420 is used to determine, for each of the first preset number of battery cell sets, a first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set, wherein the first preset number of battery cell sets are pre-determined based on a first training dataset, and the first training dataset includes the indicator data of a second preset number of sample battery cells; the sorting module 430 is used to sort the first battery cell to a target battery cell set, wherein the target battery cell set is the battery cell set with the smallest first distance among the first preset number of battery cell sets.

[0070] In one implementation, the indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time. The determining module 420 determines a first distance between the first battery cell and the cluster center based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set. This includes: determining an open-circuit voltage distance between the first battery cell and the cluster center based on the open-circuit voltage of the first battery cell and the open-circuit voltage of the cluster center of the battery cell set; determining a K-value distance between the first battery cell and the cluster center based on the K-value of the first battery cell and the K-value of the cluster center; determining an AC internal resistance distance between the first battery cell and the cluster center based on the AC internal resistance of the first battery cell and the AC internal resistance of the cluster center; determining a storage time distance between the first battery cell and the cluster center based on the storage time of the first battery cell and the storage time of the cluster center; and determining a first distance between the first battery cell and the cluster center based on the open-circuit voltage distance, the K-value distance, the AC internal resistance distance, and the storage time distance.

[0071] In one implementation, the above apparatus further includes: an update module, configured to update the cluster center of the target cell set based on the updated index data of each cell in the target cell set after the first cell is sorted into the target cell set; the acquisition module 410 is further configured to acquire the index data of the second cell; and a return module, configured to use the second cell as the first cell and return to the step of determining the first distance between the first cell and the cluster center for each of the first preset number of cell sets, based on the index data of the first cell and the index data of the cluster center of the cell set.

[0072] In one implementation, the sorting module 430 sorts the first battery cell to a target battery cell set, including: for each of the last three preset number of target battery cells added to the target battery cell set, determining a second distance between the first battery cell and the target battery cell based on the indicator data of the first battery cell and the indicator data of the target battery cell, wherein the third preset number is greater than the number of battery cells required for the battery; and sorting the first battery cell to the target battery cell set if none of the third preset number of second distances is greater than a preset threshold.

[0073] In one implementation, the indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time; the second distance includes open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance; the sorting module 430 sorts the first cell to the target cell set when the third preset number of the second distances are all not greater than a preset threshold, including: sorting the first cell to the target cell set when the third preset number of the open-circuit voltage distances are all not greater than a first threshold, the third preset number of the K-value distances are all not greater than a second threshold, the third preset number of the AC internal resistance distances are all not greater than a third threshold, and the third preset number of the storage time distances are all not greater than a fourth threshold.

[0074] In one implementation, the above apparatus further includes: the acquisition module 410, which is further configured to acquire the first training dataset before acquiring the indicator data of the first battery cell; the determination module 420, which is further configured to perform clustering based on the indicator data of the sample battery cells in the first training dataset to determine the first preset number of initial battery cell sets; the determination module 420, which is further configured to filter out sample battery cells in the target initial battery cell set from the first training dataset to determine the second training dataset, wherein the target initial battery cell set is the initial battery cell set in the first preset number of initial battery cell sets where the number of battery cells is less than a fifth threshold; and an iteration module, which is configured to perform iterative clustering based on the second training dataset until a stopping iteration condition is met to determine the first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, wherein the preset condition includes the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0075] In one implementation, the above apparatus further includes: the acquisition module 410, which is further configured to acquire the first training dataset before acquiring the indicator data of the first battery cell; the determination module 420, which is further configured to determine, for each sample battery cell in the first training dataset, the average distance between the sample battery cell and the other sample battery cells in the first training dataset based on the indicator data of the sample battery cell and the indicator data of other sample battery cells in the first training dataset; the determination module 420, which is further configured to filter out target sample battery cells from the first training dataset to determine a second training dataset, wherein the target sample battery cells are sample battery cells whose average distance is greater than a seventh threshold; and an iteration module, which is configured to perform iterative clustering based on the second training dataset until a stopping iteration condition is met to determine a first preset number of battery cell sets, wherein the stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of battery cell sets meeting a preset condition, the preset condition including the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

[0076] In one implementation, the above apparatus further includes a processing module, configured to normalize the index data of the sample cells in the second training dataset based on the maximum and minimum values ​​of the index data of the sample cells in the second training dataset before performing iterative clustering based on the second training dataset.

[0077] Optionally, such as Figure 5 As shown, this application embodiment also provides an electronic device 500, including a processor 501 and a memory 502. The memory 502 stores a program or instructions that can run on the processor 501. When the program or instructions are executed by the processor 501, they implement the various steps of the above-described battery cell sorting method embodiment and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0078] It should be noted that the electronic devices in the embodiments of this application include mobile electronic devices and non-mobile electronic devices.

[0079] This application also provides a computer-readable storage medium storing a computer-executable program or instructions. When the computer executes the program or instructions, it implements the various processes of the above-described battery cell sorting method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0080] The computer-readable storage medium may be a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0081] This application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the steps of the cell sorting method described above.

[0082] The above embodiments of this application focus on describing the differences between the various embodiments. As long as the different optimization features between the various embodiments are not contradictory, they can be combined to form a better embodiment. For the sake of brevity, they will not be described in detail here.

[0083] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for sorting battery cells, characterized in that, include: Obtain the performance data of the first battery cell; For each set of battery cells in a first preset number of battery cell sets, a first distance between the first battery cell and the cluster center is determined based on the index data of the first battery cell and the index data of the cluster center of the battery cell set. The first preset number of battery cell sets are pre-determined based on a first training dataset, and the first training dataset includes the index data of a second preset number of sample battery cells. The first battery cell is sorted into a target battery cell set, wherein the target battery cell set is the battery cell set with the smallest first distance among the first preset number of battery cell sets.

2. The method according to claim 1, characterized in that, The indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time. Determining the first distance between the first cell and the cluster center based on the indicator data of the first cell and the indicator data of the cluster center of the cell set includes: Based on the open-circuit voltage of the first cell and the open-circuit voltage of the cluster center of the cell set, the open-circuit voltage distance between the first cell and the cluster center is determined. Based on the K value of the first battery cell and the K value of the cluster center, determine the K-value distance between the first battery cell and the cluster center; Based on the AC internal resistance of the first battery cell and the AC internal resistance of the cluster center, the AC internal resistance distance between the first battery cell and the cluster center is determined. Based on the storage time of the first battery cell and the storage time of the cluster center, the storage time distance between the first battery cell and the cluster center is determined; Based on the open-circuit voltage distance, the K-value distance, the AC internal resistance distance, and the storage time distance, a first distance between the first cell and the cluster center is determined.

3. The method according to claim 1, characterized in that, After sorting the first battery cell into the target battery cell set, the method further includes: Based on the updated index data of each cell in the target cell set, the cluster center of the target cell set is updated; Obtain the aforementioned indicator data for the second battery cell; Using the second cell as the first cell, return to the execution of the step of determining the first distance between the first cell and the cluster center for each of the first preset number of cell sets, based on the index data of the first cell and the index data of the cluster center of the cell set.

4. The method according to claim 1, characterized in that, The step of sorting the first battery cell into the target battery cell set includes: For each of the third preset number of target cells added last in the target cell set, a second distance between the first cell and the target cell is determined based on the index data of the first cell and the index data of the target cell, wherein the third preset number is greater than the number of cells required for the battery. If none of the third preset number of second distances are greater than a preset threshold, the first cell is sorted into the target cell set.

5. The method according to claim 4, characterized in that, The indicator data includes open-circuit voltage, K-value, AC internal resistance, and storage time; the second distance includes open-circuit voltage distance, K-value distance, AC internal resistance distance, and storage time distance; the step of sorting the first cell into the target cell set when all of the third preset number of second distances are not greater than a preset threshold includes: When the distances between the third preset number of open-circuit voltages are all no greater than the first threshold, the distances between the third preset number of K values ​​are all no greater than the second threshold, the distances between the third preset number of AC internal resistances are all no greater than the third threshold, and the distances between the third preset number of storage times are all no greater than the fourth threshold, the first battery cell is sorted into the target battery cell set.

6. The method according to claim 1, characterized in that, Before obtaining the indicator data of the first battery cell, the method further includes: Obtain the first training dataset; Clustering is performed on the indicator data of the sample cells in the first training dataset to determine the first preset number of initial cell sets; The sample cells in the target initial cell set are filtered out from the first training dataset to determine the second training dataset, wherein the target initial cell set is the initial cell set in the first preset number of initial cell sets in which the number of cells is less than a fifth threshold; Iterative clustering is performed based on the second training dataset until the stopping iteration condition is met, and the first preset number of cell sets are determined. The stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of cell sets meeting a preset condition. The preset condition includes the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

7. The method according to claim 1, characterized in that, Before obtaining the indicator data of the first battery cell, the method further includes: Obtain the first training dataset; For each sample cell in the first training dataset, based on the index data of the sample cell and the index data of other sample cells in the first training dataset, the average distance between the sample cell and the other sample cells is determined. Target sample cells are filtered out from the first training dataset to determine the second training dataset, wherein the target sample cells are sample cells whose average distance is greater than the seventh threshold; Iterative clustering is performed based on the second training dataset until the stopping iteration condition is met, and the first preset number of cell sets are determined. The stopping iteration condition includes meeting a preset number of iterations or the cluster centers of the first preset number of cell sets meeting a preset condition. The preset condition includes the difference between cluster centers for a consecutive preset number of iterations being less than a sixth threshold.

8. The method according to claim 6 or 7, characterized in that, Before performing iterative clustering based on the second training dataset, the method further includes: Based on the maximum and minimum values ​​of the indicator data of the sample cells in the second training dataset, the indicator data of the sample cells in the second training dataset are normalized.

9. A battery cell sorting device, characterized in that, include: The acquisition module is used to acquire the indicator data of the first battery cell; The determining module is used to determine a first distance between the first battery cell and the cluster center for each battery cell set in a first preset number of battery cell sets, based on the indicator data of the first battery cell and the indicator data of the cluster center of the battery cell set, wherein the first preset number of battery cell sets are determined in advance based on a first training dataset, and the first training dataset includes the indicator data of a second preset number of sample battery cells. The sorting module is used to sort the first battery cell to a target battery cell set, wherein the target battery cell set is the battery cell set with the smallest first distance among the first preset number of battery cell sets.

10. An electronic device, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the steps of the cell sorting method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer-executable program or instructions, which, when executed by a computer, implement the steps of the cell sorting method as described in any one of claims 1-8.

Images