Battery cell selecting method and battery cell handling device

Abstract

To achieve shortening of a read time, appropriate determination of reuse of a battery cell, and appropriate maintenance management of a battery cell.SOLUTION: Regularity information among respective data regarding non-defective or defective battery cells is extracted from a plurality of past data regarding non-defective or defective battery cells in respective steps, and the regularity information among the respective data regarding the non-defective or defective battery cells is input to extract common points of the non-defective or defective battery cells and singular points of the non-defective and defective battery cells. Then, a determination model based on a conditional expression that can discriminate a non-defective battery cell and a defective battery cell from each other on the basis of the singular points is used to collect data necessary for determination using the determination model from data of the battery cells obtained in the respective steps of a manufacturing process, etc., and each management process, and the determination model is used to determine whether a battery cell is non-defective or defective.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for sorting battery cells and a battery cell handling device.

Background Art

[0002] Regarding the characteristics of battery cells for long-term use, for example, like a cycle charge-discharge test in which charge and discharge are repeatedly performed on the manufactured battery cells, when 3000 cycles (one cycle is defined as performing charge and discharge on the battery cell once each) are carried out, if the discharge capacity at that time is 80% or more compared to the capacity at the initial stage of manufacturing, it is determined to be a good product.

[0003] Also, like an aging test, the charged battery cells are left for a long time, and based on the change in voltage, the presence or absence of internal short circuit or the like of the battery cells is determined.

[0004] However, a considerable amount of time is required to conduct these tests. Therefore, instead of performing these tests on all battery cells, for each manufacturing lot, the above tests and the like are performed on randomly sampled battery cells. When the result is determined to be a good product, the other battery cells in the same lot are also considered to be good products for management.

[0005] However, in the above sorting method, since the evaluation of all the battery cells in the lot depends on the determination result of the sampled battery cells, it cannot be said to be sufficient in terms of reliability.

[0006] In addition, a single battery cell determined to be a good product in the manufacturing process or a battery module (also referred to as an assembled battery) formed by combining a plurality of these battery cells is mounted on the device as a power storage unit of the power storage device, and is connected to a charge / discharge control unit that controls the charge and discharge of the battery cell, or a CMU (Cell Management Unit: which monitors and manages so that each battery cell of the power storage unit does not enter an abnormal state, including the case of a "Cell Monitoring Unit" that only has the function of monitoring the voltage and temperature of the battery cell) or a BMU (Battery Management Unit: which monitors and manages so that the battery cells constituting the battery module of the power storage unit do not enter an abnormal state). In this connected state, while performing charge / discharge control, based on the data obtained by measuring the battery voltage and temperature of the battery cell, abnormal detection and life estimation of the battery are known to be performed.

[0007] In recent years, various predictions by machine learning, especially by organizing a large amount of information obtained by the improvement of artificial intelligence, so-called AI (Artificial Intelligence) technology, and analyzing these information by machine learning to find regularities, and using the found regularities in sales strategies, advertising strategies, etc. have been carried out. However, attempts to utilize the regularities obtained by such AI-based analysis have also been studied for the manufacturing and sales of battery cells.

[0008] As an example of technology using AI, there are technologies for quickly performing defective determination of a battery by predicting long-term characteristics using the initial characteristics of the battery (for example, see Patent Document 1), and technologies that enable early life prediction from the life trends according to various characteristic factors of a battery cell before manufacturing the battery cell to evaluate its life and improve the reliability of life prediction (for example, see Patent Document 2).

Prior Art Documents

Patent Documents

[0009]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0010] As described above, due to the progress of AI technology, it has been proposed to find regularities and laws by machine learning from various data obtained by measurement or the like, and classify and separate the object to be measured based on the regularity. In particular, by using a large variety of and large amounts of digitized data called big data, it has been realized to classify the data and find regularities from it. In such machine learning, it is possible to automatically extract feature points by deep learning (deep learning) without manually extracting them, and it has become possible to expect the extraction of feature points that are not noticed manually.

[0011] However, although the above-described technology can be applied to the life prediction of battery cells, there is room for improvement in applying it as a management index in the production process, monitoring process, and management process of battery cells.

[0012] Therefore, the present invention has been made in view of the above problems, and derives the regularity of good products and defective products in each process of battery cells from data obtained by measurement or the like, and extracts the singular points between good products and defective products of battery cells based on this regularity to generate a determination model. By using the determination model with the data of battery cells in each process as input data and performing arithmetic processing, it is possible to appropriately determine good products and defective products in each process of battery cells, realize shortening of lead time, appropriate judgment of reuse of battery cells, and appropriate maintenance management of battery cells, and an object thereof is to provide a sorting method for battery cells and a battery cell handling device.

Means for Solving the Problems

[0013] Embodiment 1; One or more embodiments of the present invention include steps of creating a determination model capable of discriminating between good battery cells and defective battery cells using data obtained in each step in the handling of battery cells having a plurality of steps; obtaining various data for each battery cell in each step; using, as input data, the data necessary for determining the battery cells using the determination model among the data obtained in each step, and performing a determination of whether the battery cells are good or defective using the determination model; and subjecting the battery cells determined to be good and the battery cells determined to be defective to different treatments according to the determination results, respectively. One of the steps is a charge / discharge step, and in the step of performing a determination of whether the battery cells are good or defective using the determination model, when the battery cells are determined to be good, a step of determining whether the battery cells need aging is provided. A method for selecting battery cells is proposed, which is characterized in that.

[0014] Embodiment 2; One or more embodiments of the present invention include a step of determining whether the battery cells need aging for exclusion determination from the production line when the battery cells are determined to be defective in the step of performing a determination of whether the battery cells are good or defective using the determination model. A method for selecting battery cells is proposed, which is characterized in that.

[0015] Embodiment 3; One or more embodiments of the present invention include a method for selecting battery cells, characterized in that the aging is a determination process for determining whether the battery cells are good or defective based on the magnitude of the self-discharge amount.

[0016] Embodiment 4; One or more embodiments of the present invention propose a method for selecting battery cells, characterized in that the determination model extracts the regular information between the data on good or defective battery cells from a plurality of past data on good or defective battery cells, extracts the specific points of the good and defective battery cells from the common points of each good or defective battery cell based on the extracted regular information, and creates a conditional expression using the data obtained in each process that can distinguish between good and defective battery cells based on the specific points.

[0017] Embodiment 5; One or more embodiments of the present invention propose a method for selecting battery cells, characterized in that the data obtained in each of the above processes is data obtained in the manufacturing process of the battery cells.

[0018] Embodiment 6; One or more embodiments of the present invention include a determination model generation unit that creates a determination model capable of distinguishing between good and defective battery cells using the data obtained in each process in the handling of battery cells having a plurality of processes, a data acquisition unit that acquires various data for each battery cell in each of the above processes, and among the data acquired in each of the above processes, using the data necessary for the determination of the battery cells using the determination model as input data, a good product determination unit that determines whether the battery cell is a good product or a defective product using the determination model, a battery cell handling unit that performs different handling on the battery cells determined to be good products and the battery cells determined to be defective products according to the determination results, and one of the above processes is a charge and discharge process, and the good product determination unit is provided with an aging determination unit that determines whether aging is required for the battery cell when the battery cell is determined to be a good product. A battery cell handling device is proposed.

[0019] Embodiment 7; One or more embodiments of the present invention propose a battery cell handling device, characterized in that when the aging determination unit determines that the battery cell is a defective product, it determines whether aging is necessary for the battery cell for the purpose of exclusion determination from the production line.

[0020] Embodiment 8; One or more embodiments of the present invention propose a battery cell handling device, characterized in that the aging is a determination process for performing a determination of whether a battery cell is a non-defective product or a defective product based on the magnitude of the self-discharge amount.

[0021] Embodiment 9; One or more embodiments of the present invention include a regularity information extraction unit that extracts regularity information between each data regarding non-defective or defective battery cells from a plurality of past data regarding non-defective or defective battery cells in each process, a storage unit that stores the regularity information between each data regarding non-defective or defective battery cells extracted by the regularity information extraction unit, a singularity extraction unit that inputs the regularity information between each data regarding the non-defective or defective battery cells from the storage unit and extracts common points for each non-defective or defective battery cell and singularities of the non-defective and defective battery cells, a conditional expression calculation unit that calculates a conditional expression capable of discriminating between non-defective battery cells and defective battery cells based on the singularities, and a determination model generation unit that verifies the validity of the conditional expression calculated by the conditional expression calculation unit and generates a determination model.

[0022] Embodiment 10; One or more embodiments of the present invention propose a battery cell handling device, characterized in that the data is data obtained in the manufacturing process of the battery cell.

Advantages of the Invention

[0023] According to one or more embodiments of the present invention, data obtained by measuring, etc., the regularity of non-defective and defective products in each process of a battery cell is derived, and based on this regularity, singular points between non-defective and defective battery cells are extracted to generate a determination model. Using the determination model by performing arithmetic processing with the data of the battery cells in each process as input data, it is possible to appropriately determine non-defective and defective products in each process of the battery cell, and there is an effect that it is possible to shorten the lead time, appropriately determine the reuse of the battery cell, and appropriately perform maintenance management of the battery cell.

Brief Description of the Drawings

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Embodiments for Carrying Out the Invention

[0025] <First Embodiment> Hereinafter, embodiments of the present invention will be described with reference to FIGS. 1 to 4.

[0026] <Electrical Configuration of Battery Cell Process Management Device> As shown in FIG. 1, the battery cell handling device 10 according to the present embodiment includes a CPU (Central Processing Unit) 100, a ROM (Read Only Memory) 200, a RAM (Random Access Memory) 300, a determination model storage unit 400, a non-defective product determination unit 500, and an interface unit 600.

[0027] The CPU 100 controls the overall processing of the battery cell handling device 10 according to a control program pre-stored in the ROM 200. In particular, in the present embodiment, in each step of the manufacturing process and the management process, based on various data obtained so far, the regularity between non-defective and defective battery cells in the manufacturing process and the management process is derived from the data obtained by measurement, etc., and based on this regularity, the singular points of non-defective and defective products are extracted, and a process of creating a determination model is performed.

[0028] The ROM 200 is a read-only non-volatile memory, and stores a control program, initial parameters, etc. as described above.

[0029] The RAM 300 is a writable and readable memory, and examples thereof include DRAM (Dynamic RAM) and SRAM (Static RAM). In the RAM 300, data and the like obtained in the process of the control process of the CPU 100 are temporarily stored.

[0030] The determination model storage unit 400 is a storage unit that stores the determination model generated by the CPU 100. The determination model to be stored in the determination model storage unit 400 may be stored in the above-described RAM 300. However, as in the present embodiment, by providing the determination model storage unit 400 that stores the determination model independently, for example, even when the RAM 300 is being used in another process, it is possible to avoid the entire arithmetic process from being delayed due to the delay in reading the determination model in the arithmetic process using the determination model.

[0031] The non-defective product determination unit 500 uses the data of the battery cells in each process as input data and performs arithmetic processing using the determination model stored in the determination model storage unit 400 to determine whether the battery cells are non-defective products or defective products. Here, the data of the battery cell refers to the data obtained during the manufacturing process, that is, the data measured in each process of the manufacturing process of the battery cell and the data acquired for production management. For example, the capacity of the battery, energy density, shape, size, material of the electrode plate, thickness of the electrode plate, length of the electrode plate, material of the electrode active material layer and the mixing ratio of the materials used, presence or absence of a binder and the type of binder, thickness of the electrode active material layer, size of the uncoated portion of the electrode active material layer, type of separator, lamination structure of the electrode plate and the separator (number of windings in the case of a wound type, number of laminations in the case of a stacked type), type of electrolyte, composition ratio of the electrolyte, presence or absence of additives such as flame retardants and their types and concentrations, drying time, drying temperature, material of the current collector, structure of the current collector, joining method and joining structure between the current collector and the electrode plate, injection time of the electrolyte, injection method of the electrolyte, time taken for impregnation and temperature during impregnation, formation charge time, formation charge voltage value, temperature during formation charge, battery voltage during charge-discharge test, battery temperature, discharge time, charge time, charge voltage, battery capacity, manufacturing lot, weather (humidity) during manufacturing, time during manufacturing, name of the manufacturing responsible person, name of the manufacturer of each material, etc. can be exemplified. In addition, information, measurement data, etc. related to the battery cell are associated with a common sorting number (manufacturing number) and stored and managed for each battery cell to be manufactured, for example, in a data server on a network. Also, into the RAM 300, information and data necessary for determination using the determination model are written from the above data server, and determination processing using the determination model is performed based on the written information (data). Note that the good product determination unit 500 may not be provided separately, and the processing of the good product determination unit 500 may be configured to be performed by the CPU 100. However, when the good product determination unit 500 is provided separately, there is an effect that the good product determination processing by the good product determination unit 500 can be executed in parallel with the processing of the CPU 100.

[0032] The interface unit 600 exchanges data with various devices connected to the network via the network. For example, in each process, when the data storage unit 1000 that stores various data obtained so far is a server or the like on a network, the battery cell handling device 10 according to the present embodiment exchanges data stored in the data storage unit 1000 via the interface unit 600.

[0033] <Functional blocks inside the CPU> As shown in FIG. 2, the CPU 100 of the battery cell handling device 10 according to the present embodiment includes a regular information extraction unit 101, a regular information storage unit 102, an outlier extraction unit 103, a conditional expression calculation unit 104, and a determination model generation unit 105.

[0034] The regular information extraction unit 101 extracts regular information regarding good or defective battery cells from a plurality of past data regarding good or defective battery cells in each process from the data storage unit 1000 provided outside, for example. Specifically, for example, using machine learning, the regularity between each data regarding good or defective battery cells may be executed by analyzing the relationship of the data. As a specific example, a method of graphing the relationship between the battery voltage, battery temperature, battery capacity and time during charging or discharging of each battery cell can be mentioned. Furthermore, after classifying the battery cells by production lot or by the type of electrolyte applied, a reclassification process is performed based on information that allows further classification. It should be noted that it is also possible by a computer to input necessary data and graph various relationships based on the input data, and it is also possible by a computer to further classify the graph by information such as the production lot of the battery cell or the type of material used, as long as a link can be established between the information regarding the battery cell and the target graph. In particular, in the present embodiment, it is important to link the obtained graph with information indicating whether the battery cell is good or defective. In addition, since unexpected common points that are not predicted may be found, graphs or distribution diagrams regarding the relationship may also be created even between information that seems to have little relevance.

[0035] The regular information storage unit 102 stores the regular information among the data related to the good or defective battery cells extracted by the regular information extraction unit 101. In this embodiment, the regular information storage unit 102 is a component of the CPU 100. However, the regular information storage unit 102 may be provided on a network as, for example, a data server.

[0036] The singularity extraction unit 103 inputs the regular information among the data related to the good or defective battery cells from the regular information storage unit 102, and extracts the common points for each of the good or defective battery cells and the singularities of the battery cells between good and defective ones. Specifically, for example, by comparing some graphed information, the common points in the good battery cells, the common points in the defective battery cells, and the differences between the good battery cells and the defective battery cells are extracted, and elements that can be the singularities that are the differences between the cases where the battery cell is good and the case where it is defective and enable the determination of good and defective are found. That is, even if it is a difference between good and defective, if that difference is not common among the good battery cells, that difference cannot be used for the determination of good and defective. Also, if the common points in the good battery cells are also common in the defective battery cells, the distinction between good and defective cannot be made. Therefore, the singularities that enable the determination of good and defective are extracted from the common points in the good battery cells, the common points in the defective battery cells, and the differences between the good battery cells and the defective battery cells.

[0037] The conditional expression calculation unit 104 calculates a conditional expression that can discriminate between good battery cells and defective battery cells based on the singularities extracted by the singularity extraction unit 103. Specifically, data required for calculating the extracted singular points is specified, an equation for calculating singular points is calculated using such data, and a conditional expression is prepared in such a form as to determine whether the numerical values regarding the singular points obtained from the calculated equation represent good products or defective products. Here, the conditional expressions to be prepared are not limited to one. If multiple singular points are found, an equation for calculating each singular point is calculated using the data required for each singular point, and conditional expressions based on the calculated equations are prepared respectively.

[0038] The determination model generation unit 105 verifies the validity of the conditional expressions calculated by the conditional expression calculation unit 104 and generates a determination model. Specifically, based on the conditional expressions calculated by the conditional expression calculation unit 104, for a plurality of battery cells that have already been determined as good products or defective products, information necessary for determining whether the battery cells are good products or defective products is applied to the conditional expressions to verify whether it is possible to correctly determine whether the battery cells are good products or defective products. If there are multiple calculated conditional expressions, verification is performed for each conditional expression respectively. As a result of the verification, if it is possible to correctly determine whether the battery cells are good products or defective products (it may be the case that all the battery cells for verifying the determination model are correctly determined as good products or defective products, or it may be the case that 95% of them can be correctly determined as good products or defective products), a model for determining good products or defective products using the conditional expression is determined as the determination model for this process. If there are multiple verified conditional expressions, as a result of the verification, select the conditional expression with the highest probability of being determined as a good product and determine the model for determining good products or defective products using it as the determination model for this process. Note that there may be multiple determination models applied to the process. For example, as a result of the verification, if multiple conditional expressions can determine good products with a similarly high probability, or if the probability of being determined as a good product in the verification result exceeds an arbitrarily set passing line probability (for example, those where 85% or more can be determined as good products), models for determining good products or defective products using each of these multiple conditional expressions may be used as the determination models respectively. A plurality of determination models for which the information necessary for determination is different will be prepared respectively. Compared with the case of one determination model, although it may become redundant because the determination of non-defective products or defective products will be performed many times, for example, when a determination model that can correctly judge 95% of non-defective products and defective products is adopted, it is expected that the remaining 5% that could not be correctly determined by this determination model can be correctly judged as non-defective products or defective products by another determination model. In addition, when making a determination with a plurality of determination models, when the determination results of X (an integer of 2 or more smaller than N) out of the determinations of N (an integer of 3 or more) determination models match, it is possible to handle non-defective products or defective products with higher accuracy by proceeding with the handling as non-defective products or defective products according to the result. In addition, priorities may be assigned in descending order of the probability of correctly judging non-defective products and defective products, the result of the determination model with the highest priority is applied with the highest priority, and within the error range of the boundary point between non-defective products and defective products in this determination model, the result of the determination model of the next priority may be adopted, and the determination may be made in multiple stages. Then, the determined determination model is output to the non-defective product determination unit 500 of the battery cell handling device 10. Note that, based on the singular points extracted by a personal computer or the like as described above, it is not limited to creating a determination model. For example, by deep learning, singular points may be automatically extracted from the input information (data), a conditional expression based on the singular points may be calculated, and the conditional expression may be verified to create a determination model.

[0039] <CPU processing> Using FIG. 3, the generation process of the determination model in the CPU 100 according to the present embodiment will be described.

[0040] The regular information extraction unit 101 inputs, from the data storage unit 1000, past data regarding defective battery cells that have been acquired so far in the target process (step S110).

[0041] Next, the Regular Information Extraction Unit 101 extracts regular information between pieces of data regarding good or defective battery cells from a plurality of past data regarding defective battery cells input from the Data Storage Unit 1000 (step S120).

[0042] The Outlier Extraction Unit 103 inputs the regular information between pieces of data regarding good or defective battery cells from the Regular Information Storage Unit 102, and extracts, for example, common points for each good or defective battery cell and outliers between good and defective battery cells using, for example, a graph of the input data classified by category (step S130).

[0043] The Conditional Expression Calculation Unit 104 calculates a conditional expression capable of discriminating between good and defective battery cells based on the outliers extracted by the Outlier Extraction Unit 103 (step S140).

[0044] The Determination Model Generation Unit 105 generates a determination model based on the conditional expression calculated by the Conditional Expression Calculation Unit 104 (step S150).

[0045] In addition, the Determination Model Generation Unit 105 applies information necessary for determining whether a battery cell is good or defective to the conditional expression for a plurality of battery cells that have already been determined to be good or defective based on the conditional expression calculated by the Conditional Expression Calculation Unit 104, and verifies whether it is possible to correctly determine whether the battery cell is good or defective (step S160).

[0046] As a result of the examination, when the Determination Model Generation Unit 105 determines that the conditional expression cannot correctly determine whether the battery cell is good or defective (''NO'' in step S160), the process returns to step S120.

[0047] On the other hand, as a result of the examination, when the Determination Model Generation Unit 105 determines that the conditional expression can correctly determine whether the battery cell is good or defective (''YES'' in step S160), the determination model is determined as the determination model in this process (step S170).

[0048] <Processing of Battery Cell Handling Device> Using FIG. 4, the processing in the battery cell handling device 10 according to the present embodiment will be described.

[0049] The CPU 100 inputs, for example, data obtained by performing processing from a data storage unit 1000, such as a server on a network, via the interface unit 600 in a target process (step S210).

[0050] The non-defective determination unit 500 performs arithmetic processing using the determination model stored in the determination model storage unit 400 with the data of the battery cells in each process as input data (step S220).

[0051] The non-defective determination unit 500 makes a non-defective determination or a defective determination of the battery cell as a result of the arithmetic processing (step S230).

[0052] The determination result is output from, for example, the interface unit 600 via the CPU 100 and is used for control of a functional unit that performs sorting processing (post-processing step) between non-defective and defective battery cells (step S240).

[0053] As described above, in the battery cell handling device 10 according to the present embodiment, information (data) acquired in each process of the battery cell is used, and from these information, characteristic points found in non-defective battery cells, characteristic points found in defective battery cells, and singularities found between non-defective and defective battery cells are extracted, and arithmetic processing is performed between the extracted characteristic points and singularities and various information (data) obtained in the actual process to generate a determination model for determining whether the battery cell is non-defective or defective. Therefore, it is possible to appropriately determine whether the battery cell is non-defective or defective in each process of the battery cell.

[0054] In addition, by deriving from data obtained through measurement or the like the regularity between non-defective and defective battery cells in each process of the battery cell, extracting the singular points of non-defective and defective battery cells, and appropriately determining non-defective or defective battery cells in each process of the battery cell, for example, in the manufacturing process, the lead time can be shortened, in the reuse process, an appropriate determination for reusing the battery cell can be provided, and in the maintenance management process, appropriate maintenance management for the battery cell can be realized.

[0055] <Second Embodiment> Hereinafter, embodiments of the present invention will be described with reference to FIGS. 5 to 7. Note that, in this embodiment, the charging and discharging process is exemplified among the production processes of the battery cell.

[0056] <Electrical Configuration of Battery Cell Handling Device> As shown in FIG. 5, the battery cell handling device 20 according to this embodiment includes a CPU (Central Processing Unit) 110, a ROM (Read Only Memory) 200, a RAM (Random Access Memory) 300, a determination model storage unit 410, a non-defective determination unit 510, an interface unit 600, and an aging determination unit 700. Note that components denoted by the same reference numerals as those in the first embodiment have the same functions, and thus detailed descriptions thereof are omitted.

[0057] The CPU 110 controls the overall processing of the battery cell handling device 10 according to a control program for the charging and discharging process stored in advance in the ROM 200. In particular, in this embodiment, in the charging and discharging process, based on various data obtained so far, the regularity between non-defective and defective battery cells in the charging and discharging process is derived from data obtained through measurement or the like, and based on this regularity, the singular points of non-defective and defective products are extracted, and a process of generating a determination model is performed.

[0058] The determination model storage unit 410 is a storage unit that stores the determination model generated by the CPU 110.

[0059] The non-defective determination unit 510 uses the determination model stored in the determination model storage unit 410 with the data of the battery cell in the charge and discharge process as input data, and performs arithmetic processing to determine whether the battery cell is non-defective or defective. Here, examples of the data of the battery cell can include, for example, the battery voltage, battery temperature, discharge time, charge time, charge voltage, etc. in the charge and discharge process.

[0060] When the battery cell is determined to be non-defective, the aging determination unit 700 determines whether aging is present. Here, aging is a process of determining whether the battery cell is truly non-defective or defective based on whether an abnormal state has occurred, that is, a state where the self-discharge amount is larger than that of a non-defective battery cell, due to causes such as an internal short circuit. This is a determination process performed by charging the battery cell to a certain extent and then leaving the battery cell for a long period, such as one week or one month.

[0061] As shown in FIG. 6, the CPU 110 includes a regular information extraction unit 111, a regular information storage unit 112, an outlier extraction unit 113, a conditional expression calculation unit 114, and a determination model generation unit 115.

[0062] The regular information extraction unit 111 extracts regular information between pieces of data regarding non-defective or defective battery cells from a plurality of past data regarding non-defective or defective battery cells in the charge and discharge process from, for example, an externally provided data storage unit 1000.

[0063] The regular information storage unit 112 stores the regular information between pieces of data regarding non-defective or defective battery cells extracted by the regular information extraction unit 111.

[0064] The outlier extraction unit 113 inputs the regular information between pieces of data regarding non-defective or defective battery cells from the regular information storage unit 112, and extracts the common points for each non-defective or defective battery cell and the outliers of the non-defective and defective battery cells.

[0065] Based on the singular points extracted by the singular point extraction unit 113, the conditional expression calculation unit 114 calculates a conditional expression capable of discriminating between good battery cells and defective battery cells.

[0066] The determination model generation unit 115 verifies the validity of the conditional expression calculated by the conditional expression calculation unit 114 and generates a determination model.

[0067] <Processing of the battery cell handling device> The processing in the battery cell handling device 10 according to the present embodiment will be described with reference to FIG. 7.

[0068] The CPU 110 inputs data obtained by performing processing via the interface unit 600 from the data storage unit 1000, which is, for example, a server on a network, in the charge and discharge process (step S310). In the present embodiment, there is one determination model, and the data required for determination by the determination model will be described as the data of the battery cell in the charge and discharge process.

[0069] The good product determination unit 510 performs arithmetic processing using the determination model stored in the determination model storage unit 410 with the data of the battery cell in the charge and discharge process as input data (step S320). Thereby, information on the charging voltage, discharging voltage, and capacity obtained in the charge and discharge process is acquired, and the charge and discharge characteristics are obtained from the relationship between the charging voltage, discharging voltage, and capacity.

[0070] The good product determination unit 510 determines whether it is possible to determine that the battery cell is a good product based on the result of the calculation by the determination model (step S330). Here, the determination as to whether the battery cell is a good product or a defective product is made based on whether the result calculated by the determination model is a numerical value within the range in which the battery cell is determined to be a good product.

[0071] When it is determined that the battery cell is not a good product (\"NO\" in step S330), the battery cell is excluded from the original production line without proceeding to the next processing step for disposal as a defective product, and disposal processing (such as carrying it into a storage for waste) is performed (step S340).

[0072] On the other hand, when it is determined by the good product determination unit 510 that the battery cell is a good product (\"YES\" in step S330), next, the good product determination unit 510 determines whether aging processing is necessary (step S350). Here, the determination of whether aging processing is necessary is, for example, as a determination result by the good product determination unit 510 in step S340, determined based on whether the battery cell is near the limit point for distinguishing between good and defective battery cells although it is determined to be a good product. Specifically, it is determined based on whether it is included within a predetermined error range with respect to the boundary point between good and defective battery cells.

[0073] Therefore, when the charge-discharge characteristics of the battery cell are determined to be a good product within a predetermined error range of the boundary point between good and defective battery cells, for example, within the range of the boundary point + (or minus) 15% (\"NO\" in step S350), the good product determination unit 510 determines that normal aging processing is necessary and makes a determination to execute normal aging processing (step S370).

[0074] On the other hand, when the charge-discharge characteristics of the battery cell are determined to be a good product outside the predetermined error range of the boundary point between good and defective battery cells, for example, outside the range of the boundary point + (or minus) 15% (\"YES\" in step S350), the good product determination unit 510 determines that normal aging processing is not necessary and makes a determination not to execute aging processing (step S360).

[0075] As described above, according to the present embodiment, the good product determination unit 510 does not determine to execute normal aging processing for all battery cells that are good products. Instead, when the charge and discharge characteristics of the battery cell are determined to be good products within a predetermined error range of the boundary point between good and defective battery cells, for example, within the range of the boundary point + (or minus) 15%, it is determined that normal aging processing is necessary. When the charge and discharge characteristics of the battery cell are determined to be good products outside the predetermined error range of the boundary point between good and defective battery cells, for example, outside the range of the boundary point + (or minus) 15%, it is determined that normal aging processing is unnecessary. Therefore, by applying the determination model to the charge and discharge process of the battery cell, the production efficiency of the battery cell can be improved.

[0076] <Modification Example> Note that in the present embodiment, in the determination step (step S330) of whether the battery cell can be determined to be a good product based on the result of the calculation by the determination model in FIG. 7, when it is determined that the battery cell is not a good product (''NO'' in step S330), disposal processing (step S340) is performed. However, even when it is determined that the product is not a good product, it may be determined whether aging processing is necessary. For example, when it is determined that the battery cell is a defective product within the error range of the boundary point between good and defective battery cells as described above, it is determined that normal aging processing is necessary. When the charge and discharge characteristics of the battery cell are determined to be defective products outside the predetermined error range of the boundary point between good and defective battery cells, disposal processing may be performed.

[0077] Specifically, as shown in FIG. 8, the good product determination unit 510 determines whether the battery cell can be determined to be a good product based on the result of the calculation by the determination model (step S330). Then, when it is determined by the good product determination unit 510 that the battery cell is not a good product (''NO'' in step S330), it is determined whether aging processing is necessary (step S380). Here, when the charge-discharge characteristics of the battery cell are determined to be defective within a predetermined error range at the boundary point between good and defective battery cells, for example, within the range of the boundary point + (or minus) 15% (the "NO" in step S380), the good-product determination unit 510 determines to execute normal aging processing on the assumption that normal aging processing is necessary (step S390).

[0078] On the other hand, when the charge-discharge characteristics of the battery cell are determined to be defective outside the predetermined error range at the boundary point between good and defective battery cells, for example, outside the range of the boundary point + (or minus) 15% (the "YES" in step S350), the good-product determination unit 510 excludes the battery cell from the original production line without proceeding to the next processing step for disposal as a defective product, and performs disposal processing (such as carrying it into a storage for waste) (step S340).

[0079] In this way, for battery cells determined to be good or defective within the error range at the boundary point between good and defective battery cells, by performing normal aging processing, the aging processing of battery cells determined not to require aging processing is omitted, and aging processing can be performed on other battery cells. Therefore, while reducing the number of battery cells to be aged, the good and defective battery cells can be more reliably sorted. Note that by feeding back the result of this aging process to the creation of the determination model and creating (or updating the created determination model) a determination model that can surely determine good and defective battery cells, the number of battery cells included in the range of the boundary point between good and defective battery cells can be reduced, and a determination model that can more accurately determine good and defective battery cells can be constructed.

[0080] In addition to updating the determination model by feeding back the results of the aging process to the creation of the determination model, when the results of the aging process are used for extracting regular information or outliers regarding good or defective battery cells in machine learning for creating the determination model (S120 and S13 in FIG. 3) and such regular information or outliers are caused by the raw material lot, rather than feeding back to the creation of the determination model, information on the regularities or outliers may be provided (fed back) to a stage prior to the production process stage, such as reviewing the incoming standards of the raw materials. Thus, by outputting the regular information and outlier information according to the nature of the regularities and outliers extracted in addition to feeding back the new determination results in the handling process of the battery cells to which the determination model is applied to the update (creation) of the determination model, it can be expected to be reflected in the improvement of the performance and quality of the battery by enabling utilization at a more appropriate stage in the handling process of the battery cells.

[0081] Also, in this embodiment, an example in which the implementation of the aging process is selectively applied to the data required for determination by the determination model as the data of the battery cells in the charge and discharge process has been described. However, when the data required for determination by the determination model is data obtained in a process other than the charge and discharge process, for example, data obtained in a process before the charge and discharge process, it may be possible to make a determination as to whether it is a good product or a defective product at a stage before the charge and discharge process, that is, at a stage when the data required for determination by the determination model is complete. It is preferable that the battery cells determined to be defective be made recognizable as defective products that do not need to undergo subsequent processes by linking them to the manufacturing numbers, etc. specified by the barcodes attached to the battery cells. By doing so, in subsequent processes, it can be managed so that battery cells recognized as defective when the barcode is read can be removed from the production line at a timing when they can be removed.

[0082] In addition, only the battery cells determined to be good (only the battery cells not determined to be defective) may be subjected to subsequent charge and discharge processes, aging processes, and the like. In this way, the CPU or the like manages whether or not the data required for the determination model is complete, and when the data required for the determination by the determination model is complete, the determination model determines whether the battery cell is good or defective. By linking the determination result of whether the battery cell is good or defective by the determination model and the information that can identify the battery cell flowing on the production line, the following advantages are obtained. The manufacturing process of the battery cell consists of a plurality of processes. In one or more of these plurality of processes, when a step of excluding defective products is included by performing a determination of whether the battery cell is good or defective without using the determination model in the present embodiment, if the determination result of whether the battery cell is good or defective by the above-described determination model that is performed in parallel separately from the determination of whether the battery cell is good or defective without using these determination models is linked to the information that can identify the battery cell, then when determining whether the battery cell is good or defective without using the determination model, by reading the information that can identify the battery cell to be determined and confirming the determination result of whether the battery cell is good or defective by the above-described determination model, by recognizing the information that the battery cell is a defective product as the determination result of whether the battery cell is good or defective by the above-described determination model, it is possible to exclude the defective battery cells without performing a determination of whether the battery cell is good or defective without using the determination model.

[0083] <Third Embodiment> Hereinafter, embodiments of the present invention will be described with reference to FIGS. 9 to 11. Note that, in this embodiment, as a process, a reuse (reuse determination) process is exemplified.

[0084] <Electrical Configuration of Battery Cell Handling Device> As shown in FIG. 8, the battery cell handling device 30 according to this embodiment includes a CPU (Central Processing Unit) 120, a ROM (Read Only Memory) 200, a RAM (Random Access Memory) 300, a determination model storage unit 420, a non-defective product determination unit 520, an interface unit 600, and a reuse determination unit 800. Note that components denoted by the same reference numerals as those in the first and second embodiments have the same functions, and thus detailed descriptions thereof are omitted.

[0085] The CPU 120 controls the processing of the entire battery cell handling device 10 according to a control program for the reuse process pre-stored in the ROM 200. In particular, in this embodiment, based on various data obtained in a reuse process including a charge / discharge process of charging and discharging the battery cells in the production process or the recovered battery cells, a process of measuring the capacity at full charge, etc., the regularity between non-defective and defective battery cells in the reuse process is derived from the data obtained by measurement or the like, and based on this regularity, the singular points of non-defective and defective products are extracted, and a process of generating a determination model is performed.

[0086] The determination model storage unit 420 is a storage unit that stores the determination model generated by the CPU 120.

[0087] The non-defective product determination unit 520 uses the determination model stored in the determination model storage unit 420 as input data with the data of the battery cells obtained in a reuse process including a charge / discharge process of charging and discharging the battery cells in the production process or the recovered battery cells, a process of measuring the capacity at full charge, etc., and performs arithmetic processing to determine whether the battery cells are non-defective or defective. Here, examples of the data of the battery cells can include information regarding battery temperature, battery voltage, battery capacity, number of charge / discharge cycles, years of use, presence or absence of a long-term unused period, number of errors, error occurrence time, error content, and the like.

[0088] When the battery cell is determined to be a non-defective product, the reuse determination unit 800 determines whether both the full charge characteristics and the temperature characteristics are within the range that enables the reuse of the battery cell. Here, reuse means using the recovered battery cell or power storage system as it is again. Even if the charge capacity at full charge has decreased due to deterioration compared to when it was delivered, it is required that charging up to the desired capacity can be ensured, and that the temperature of the battery cell does not increase due to charge and discharge. In the following, in particular, the case of using a determination model that can classify battery cells into non-defective and defective products based on the relationship between the charge and discharge voltage and capacity of the battery cell (charge and discharge characteristics), the capacity value at full charge (full charge characteristics), and the relationship between the charge and discharge voltage and the temperature change of the battery cell (temperature characteristics) will be described.

[0089] Also, as shown in FIG. 9, the CPU 120 includes a regular information extraction unit 121, a regular information storage unit 122, an outlier extraction unit 123, a conditional expression calculation unit 124, and a determination model generation unit 125.

[0090] The regular information extraction unit 121 extracts regular information between each piece of data regarding non-defective or defective battery cells from a plurality of past data regarding non-defective or defective battery cells obtained in a reuse process including processes such as charge and discharge processing for performing charge and discharge on the production process or the recovered battery cells from an externally provided data storage unit 1000, and capacity measurement at full charge.

[0091] The regular information storage unit 122 stores the regular information between each piece of data regarding non-defective or defective battery cells extracted by the regular information extraction unit 121.

[0092] The outlier extraction unit 123 inputs the regular information between each piece of data regarding non-defective or defective battery cells from the regular information storage unit 122, and extracts the common points for each non-defective or defective battery cell and the outliers of the non-defective and defective battery cells.

[0093] Based on the singular points extracted by the singular point extraction unit 123, the conditional expression calculation unit 124 calculates a conditional expression capable of discriminating between good battery cells and defective battery cells.

[0094] The determination model generation unit 125 verifies the validity of the conditional expression calculated by the conditional expression calculation unit 124 and generates a determination model.

[0095] <Processing of the battery cell handling device> With reference to FIG. 10, the processing in the battery cell handling device 30 according to the present embodiment will be described.

[0096] The CPU 120 inputs data obtained in the reuse process including a charge and discharge process for charging and discharging the battery cells in the production process or the recovered battery cells, a capacity measurement at full charge, etc., from the data storage unit 1000 such as a data server via the interface unit 600 (step S410).

[0097] Using the battery cell data obtained in the reuse process including a charge and discharge process for charging and discharging the battery cells in the production process or the recovered battery cells, a capacity measurement at full charge, etc., as input data, the good product determination unit 520 performs arithmetic processing using the determination model stored in the determination model storage unit 420 (step S420). Thereby, the relationship between the charge and discharge voltage and the capacity of the battery cell (charge and discharge characteristics), the capacity value at full charge (full charge characteristics), and the relationship between the charge and discharge voltage and the temperature change of the battery cell (temperature characteristics) are obtained.

[0098] Based on the result of the calculation by the determination model, the good product determination unit 520 determines whether it is possible to determine that the battery cell is a good product (step S430). Here, the determination as to whether the battery cell is a good product or a defective product is made, for example, based on whether the result calculated by the determination model based on the charge and discharge characteristics is a numerical value within the range in which the battery cell is determined to be a good product.

[0099] When it is determined that the battery cell is not a non-defective product ( "NO" in step S430), the battery cell is removed from the original production line without proceeding to the next processing step for disposal as a defective product, and disposal processing (such as carrying it into a storage for waste) is performed (step S440).

[0100] On the other hand, when the non-defective product determination unit 520 determines that the battery cell is a non-defective product ( "YES" in step S430), next, the reuse determination unit 800 determines whether the battery cell is reusable (step S450). Here, the determination of whether reuse is possible is made based on, for example, the full charge characteristics and temperature characteristics.

[0101] When the reuse determination unit 800 determines that at least one of the full charge characteristics or temperature characteristics of the battery cell is outside the range determined to be reusable ( "NO" in step S450), the battery cell is disposed of as a defective product that cannot be reused (step S440).

[0102] On the other hand, when the reuse determination unit 800 determines that both the full charge characteristics and temperature characteristics are within the range determined to be reusable ( "YES" in step S450), it is reused as a battery cell and its state is managed (step S460).

[0103] As described above, according to the present embodiment, the reuse determination unit 800 determines whether the battery cell is reusable based on whether at least one of the full charge characteristics or temperature characteristics of the battery cell determined to be a non-defective product by the non-defective product determination unit 520 is within the range determined to be reusable. Therefore, by applying the determination model to the reuse process of the battery cell, the battery cell can be reused efficiently and appropriately.

[0104] <Fourth Embodiment> Hereinafter, embodiments of the present invention will be described with reference to FIGS. 12 to 14. Note that this embodiment exemplifies, as a process, a monitoring process of monitoring the behavior of a battery cell mounted on and operating in a power storage system, for example.

[0105] <Electrical Configuration of Battery Cell Handling Device> As shown in FIG. 11, the battery cell handling device 40 according to this embodiment includes a CPU (Central Processing Unit) 130, a ROM (Read Only Memory) 200, a RAM (Random Access Memory) 300, a determination model storage unit 430, a non-defective product determination unit 530, an interface unit 600, and a life prediction unit 900. Note that components denoted by the same reference numerals as those in the first embodiment, the second embodiment, and the third embodiment have the same functions, and thus detailed descriptions thereof are omitted.

[0106] The CPU 130 controls the processing of the entire battery cell handling device 10 according to a control program for the monitoring process stored in advance in the ROM 200. In particular, in this embodiment, based on various data obtained in the production process or the monitoring process, the regularity between non-defective and defective battery cells in the monitoring process is derived from the data obtained by measurement or the like, and based on this regularity, the singular points of non-defective and defective products are extracted, and a process of generating a determination model is performed.

[0107] The determination model storage unit 430 is a storage element that stores the determination model generated by the CPU 130.

[0108] The non-defective product determination unit 530 uses the determination model stored in the determination model storage unit 430 with the data of the battery cell obtained in the production process or the monitoring process as input data, and performs arithmetic processing to determine whether the battery cell is a non-defective product or a defective product. Here, examples of the data of the battery cell can include information such as battery temperature, battery voltage, battery capacity, number of charge and discharge cycles, years of use, presence or absence of a long-term unused period, number of errors, error occurrence time, and error content.

[0109] When the battery cell is determined to be a non-defective product, the life prediction unit 900 predicts the life of the battery cell. Here, the life prediction is made when the battery voltage or battery temperature of the battery cell measured in the monitoring process is a voltage value or temperature in a range that cannot be measured in the case of a normal battery cell. For a battery cell determined not to be normally usable as a battery cell, a warning is given to inform of the prohibition of continued use. In the actual monitoring process, it can be determined whether the battery voltage or battery temperature of the battery cell to be monitored is within the normal value range. In the present embodiment, even for those determined to be normal as long as the battery voltage or battery temperature of the monitored battery cell is concerned, a state indicating a sign that the life of the battery cell is approaching (the deterioration of the battery cell is progressing) is modeled by a determination model and judged by calculation from the measured data. Hereinafter, a determination model capable of making a deterioration determination will be exemplified and described using information on the battery voltage and battery temperature of the battery cell measured in the battery cell monitoring process.

[0110] Further, as shown in FIG. 12, the CPU 130 includes a regular information extraction unit 131, a regular information storage unit 132, an outlier extraction unit 133, a conditional expression calculation unit 134, and a determination model generation unit 135.

[0111] The regular information extraction unit 131 extracts regular information between pieces of data on non-defective or defective battery cells from a plurality of past data on non-defective or defective battery cells obtained in the production process or the monitoring process from, for example, an externally provided data storage unit 1000.

[0112] The regular information storage unit 132 stores the regular information between pieces of data on non-defective or defective battery cells extracted by the regular information extraction unit 131.

[0113] The singularity extraction unit 133 inputs the regular information between the data regarding the good or defective battery cells from the regular information storage unit 132, and extracts the common points for each good or defective battery cell and the singularities of the good and defective battery cells.

[0114] Based on the singularities extracted by the singularity extraction unit 133, the conditional expression calculation unit 134 calculates a conditional expression capable of discriminating between good battery cells and defective battery cells.

[0115] The determination model generation unit 135 verifies the validity of the conditional expression calculated by the conditional expression calculation unit 134 and generates a determination model.

[0116] <Processing of the battery cell handling device> With reference to FIG. 13, the processing in the battery cell handling device 40 according to the present embodiment will be described.

[0117] The CPU 130 inputs data such as the battery voltage and battery temperature of the battery cell obtained in the production process or the monitoring process from the data storage unit 1000, such as a data server, via the interface unit 600 (step S510).

[0118] The good product determination unit 530 uses the determination model stored in the determination model storage unit 420 with the data of the battery cell such as the battery voltage and battery temperature of the battery cell obtained in the production process or the monitoring process as input data, and performs arithmetic processing (step S520). Thereby, the relationship (temperature characteristic) between the charge and discharge voltage of the battery cell and the temperature change is obtained.

[0119] Based on the result of the calculation by the determination model, the good product determination unit 530 determines whether or not it is possible to determine that the battery cell is a good product (normal) (step S530). Here, the determination as to whether the battery cell is a good product or a defective product is made, for example, based on whether the result calculated by the determination model based on the temperature characteristic is a numerical value within the range in which the battery cell is determined to be a good product.

[0120] When it is determined that the battery cell is not a good product (abnormal) ( "NO" in step S530), it is assumed that there is some abnormality in the battery cell, and the user of the power storage system in which the battery cell is mounted is notified of the abnormality of the battery cell (step S540). At this time, for safety reasons, the operation of the power storage system itself may be prohibited by remote operation.

[0121] On the other hand, when it is determined by the good product determination unit 520 that the battery cell is a good product (normal) ( "YES" in step S530), next, the life prediction unit 900 determines whether or not the battery cell is deteriorated (step S550). Here, the determination as to whether or not the battery cell is deteriorated can be made based on the calculation result obtained by calculating the battery voltage and the battery temperature measured using the determination model.

[0122] When it is determined by the life prediction unit 900 that there are signs of deterioration in the battery cell and it is determined that the life of the battery cell is short ( "NO" in step S550), the user is warned to that effect, and the battery cell is monitored more intensively. For example, measures such as replacing the battery cell in advance are promoted so that the battery cell does not deteriorate rapidly and cannot exhibit sufficient functions during a power outage or the like (step S570).

[0123] On the other hand, when it is determined that there are no signs of deterioration in the battery cell ( "YES" in step S550), in particular, the monitoring of the battery cell is continued without giving a warning to the user or the like (step S560).

[0124] As described above, according to the present embodiment, the life prediction unit 900 determines whether or not the battery voltage and the battery temperature of the battery cell being monitored are within the normal value range in the monitoring process among the battery cells determined to be good products by the good product determination unit 530.

[0125] In addition, predict the life of the battery cell, and for those with a short life, warn the user to that effect, and strengthen the monitoring of the battery cell, etc., so that the battery cell does not rapidly deteriorate and cannot perform sufficient functions during a power outage or the like, and promote countermeasures such as replacing the battery cell in advance. Therefore, by applying the determination model to the monitoring process of the battery cell, the monitoring of the battery cell can be performed more precisely.

[0126] In addition, in the above embodiment, a determination model was created, and various selections (selection of good or defective products, selection of whether reuse is possible, selection of the degree of monitoring) using this determination model were also exemplified, but it is not limited to this.

[0127] For example, as described above, there may be a plurality of determination models to be created instead of just one.

[0128] For example, when a plurality of determination models with different required information for determination are created, they may be used as appropriate, and after the process steps where determination is possible, determination by each determination model may be performed, and selection processing may be performed as appropriate each time.

[0129] Also, for example, if the determination of whether a battery cell is a good or defective product is performed by a plurality of determination models and selection processing is executed, and this is done for each appropriate processing step, defective battery cells can be removed as appropriate in each processing step, and the efficiency of the processing can be improved without performing unnecessary processing on defective battery cells.

[0130] Also, for example, if one determination model can determine that a battery cell is defective with a probability of 90%, and by using another determination model that uses information different from the information required by one determination model, it is possible to determine that the remaining 10% of the battery cells are defective with a probability of about 90%, after performing the selection processing of whether a battery cell is a good or defective product by one determination model, the selection processing of whether a battery cell is a good or defective product may be performed on the battery cells that could not be determined to be defective by another determination model. If different determination models are prepared and selection processing is performed in two or more steps in this way, defective battery cells can be extracted with higher accuracy, and the efficiency of processing for defective battery cells can be improved.

[0131] Note that the battery cell sorting method can be recorded on a computer-readable recording medium, and the battery cell handling device of the present invention can be realized by having the program recorded on this recording medium read and executed by a battery cell handling device or the like. The computer system referred to here includes hardware such as an OS and peripheral devices.

[0132] Also, the "computer system" shall include a homepage providing environment (or display environment) if the WWW (World Wide Web) system is used. Further, the above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) like the Internet or a communication line (communication wire) like a telephone line.

[0133] Also, the above program may be for realizing a part of the functions described above. Furthermore, it may be a so-called difference file (difference program) that can realize the functions described above in combination with a program already recorded in the computer system.

[0134] As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and designs and the like within the scope not departing from the gist of the present invention are also included.

Explanation of Reference Numerals

[0135] 10; Battery cell handling device 20; Battery cell handling device 30; Battery cell handling device 40; Battery cell handling device 100; CPU 101; Regular information extraction unit 102; Regular information storage unit 103; Singularity extraction unit 104; Conditional expression calculation unit 105; Judgment model generation unit 110; CPU 111; Regular information extraction unit 112; Regular information storage unit 113; Singularity extraction unit 114; Conditional expression calculation unit 115; Judgment model generation unit 120; CPU 121; Regular information extraction unit 122; Regular information storage unit 123; Singularity extraction unit 124; Conditional expression calculation unit 125; Judgment model generation unit 130; CPU 131; Regular information extraction unit 132; Regular information storage unit 133; Singularity extraction unit 134; Conditional expression calculation unit 135; Judgment model generation unit 200; ROM 300; RAM 400; Judgment model storage unit 410; Judgment model storage unit 420; Judgment model storage unit 430; Judgment model storage unit 500; Good product judgment unit 510; Good product judgment unit 520; Good product judgment unit 530; Good product judgment unit 600; Interface unit 700; Aging judgment unit 800; Reuse judgment unit 900; Life prediction unit 1000; Data storage unit

Claims

1. A step of creating a determination model capable of discriminating between the good battery cells and the defective battery cells using data obtained in each process in the handling of battery cells having a plurality of processes; A step of obtaining various data for each battery cell in each of the above processes; A step of using, as input data, data necessary for the determination of the battery cells using the determination model among the data obtained in each of the above processes, and performing a good product determination or a defective product determination of the battery cells using the determination model; A step of performing different treatments according to the determination results for the battery cells determined to be good products and the battery cells determined to be defective products, respectively, as a result of the determination; Including, One of the above processes is a recovery process, In the step of performing a good product determination or a defective product determination of the battery cells using the determination model, when the battery cells are determined to be good products, a step of determining whether both the capacity value and the temperature characteristics at full charge are within a reusable range is provided. A method for sorting battery cells, characterized in that.

2. In the step of performing a good product determination or a defective product determination of the battery cells using the determination model, when the battery cells are determined to be defective products, a step of targeting the battery cells for disposal is provided. The method for sorting battery cells according to claim 1, characterized in that.

3. The determination model extracts regular information between each data regarding the good or defective battery cells from a plurality of past data regarding the good or defective battery cells, and extracts the singularities of the good and defective battery cells from the common points for each of the good or defective battery cells based on the extracted regular information, and uses the data obtained in each process to create a conditional expression capable of discriminating between the good battery cells and the defective battery cells based on the singularities. The method for sorting battery cells according to claim 1 or claim 2, characterized in that.

4. The data obtained in each of the above processes is data obtained in the manufacturing process of the battery cells. The method for sorting battery cells according to claim 1 or claim 2, characterized in that.

5. A determination model generation unit that creates a determination model capable of discriminating between the good battery cells and the defective battery cells using data obtained in each process in the handling of battery cells having a plurality of processes; In each of the above steps, a data acquisition unit that acquires various data for each battery cell; Among the data acquired in each of the above steps, using the determination model, a non-defective determination unit that performs non-defective or defective determination of the battery cell using, as input data, the data necessary for the determination of the battery cell using the determination model; A battery cell handling unit that performs different handling according to the determination result on the battery cell determined to be non-defective and the battery cell determined to be defective as a result of the determination; comprising; One of the above steps is a recovery step, In the step of performing non-defective or defective determination of the battery cell using the determination model, when the battery cell is determined to be non-defective, a reuse determination unit that determines whether both the capacity value and the temperature characteristics at full charge are within a reusable range, characterized in that the battery cell handling device is provided with the reuse determination unit. **Claim 6** The reuse determination unit determines whether the battery cell determined to be non-defective by the non-defective determination unit is reusable, and a battery cell determined not to be reusable as a result of the determination is to be discarded, characterized in that the battery cell handling device according to claim 5 is provided. **Claim 7** A regular information extraction unit that extracts regular information between each data regarding the non-defective or defective battery cell from a plurality of past data regarding the non-defective or defective battery cell in each of the above steps; A storage unit that stores the regular information between each data regarding the non-defective or defective battery cell extracted by the regular information extraction unit; A specific point extraction unit that inputs the regular information between each data regarding the non-defective or defective battery cell from the storage unit and extracts common points for each non-defective or defective battery cell and specific points of the non-defective and defective battery cells; A conditional expression calculation unit that calculates a conditional expression that can discriminate between the non-defective battery cell and the defective battery cell based on the specific points; A determination model generation unit that verifies the validity of the conditional expression calculated by the conditional expression calculation unit and generates the determination model; characterized in that it includes the battery cell handling device according to claim 5 or claim 6. **Claim 8** The data is data obtained in the manufacturing process of the battery cell, characterized in that the battery cell handling device according to claim 5 or claim 6 is provided.

Images