Battery diagnostic device and its operating method

Abstract

According to embodiments disclosed in this document, the battery diagnostic device may include: an information acquisition unit that acquires characteristic data for each battery cell; and a controller that extracts latent variables from the characteristic data, makes a primary determination of whether each battery cell is defective based on the feature values ​​corresponding to the latent variables, and makes a secondary determination of whether each battery cell is defective based on the primary determination result and the distribution of the feature values.

Description

Technical Field

[0001] The present invention claims the benefit of priority based on Korean Patent Application No. 10-2023-0097064 filed on July 25, 2023, and all the contents disclosed in the document of the Korean patent application are incorporated herein by reference in their entirety. The embodiments disclosed in this document relate to a battery diagnostic device and an operating method thereof.

Background Art

[0002] In recent years, research and development on secondary batteries have been actively conducted. Here, a secondary battery is a battery that can be charged and discharged, and includes both conventional Ni / Cd batteries, Ni / MH batteries, etc., and recent lithium-ion batteries. Among secondary batteries, lithium-ion batteries have the advantage of having a much higher energy density compared to conventional Ni / Cd batteries, Ni / MH batteries, etc. In addition, since lithium-ion batteries can be manufactured in a small and lightweight manner, they are used as a power source for mobile devices, and in recent years, their usage range has been extended to the power source of electric vehicles and they have attracted attention as a next-generation energy storage medium.

[0003] When battery defects such as disconnection, negative electrode exposure, and lithium precipitation occur in the battery cells included in a battery pack, there is a risk of fire, so it is necessary to detect defective battery cells early and take measures before a fire occurs.

Summary of the Invention

Problems to be Solved by the Invention

[0004] One object of the embodiments disclosed in this document is to provide a battery diagnostic device and an operating method thereof that can detect battery cell defects early and prevent the risk of fire.

[0005] The technical problems of the embodiments disclosed in this document are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0006] According to embodiments disclosed in this document, the battery diagnostic device may include: an information acquisition unit that acquires characteristic data for each battery cell; and a controller that extracts latent variables from the characteristic data, makes a primary determination of whether each battery cell is defective based on the feature values ​​corresponding to the latent variables, and makes a secondary determination of whether each battery cell is defective based on the primary determination result and the distribution of the feature values.

[0007] According to the embodiment, the controller can acquire a distribution diagram showing the distribution of the feature values ​​and make a secondary determination of whether each of the battery cells is defective based on the position of each battery cell on the distribution diagram.

[0008] According to the embodiment, the controller can, as a result of the initial determination, make a secondary determination that cells that are determined to be normal as a result of the initial determination and whose distance from other cells that are determined to be normal on the distribution diagram is at or above a first level are defective, and can make a secondary determination that cells that are determined to be defective as a result of the initial determination and whose density from other cells that are determined to be defective on the distribution diagram is at or above a second level are normal.

[0009] According to the embodiment, the controller can input the characteristic data into an autoencoder that has already been trained to extract the latent variables. According to the embodiment, the autoencoder can learn the process of extracting the latent variables by receiving characteristic data relating to a normal cell as input. According to the embodiment, the controller can input the feature values ​​into a model that has already been trained to obtain the initial judgment result.

[0010] According to the embodiment, the characteristic data may include differential capacitance data (dQ / dV) of the battery. According to the embodiment, the information acquisition unit can acquire the characteristic data during the low-current interval at the initial stage of activation of the battery cell.

[0011] According to embodiments disclosed herein, a battery diagnostic method may include the steps of: acquiring characteristic data for each battery cell; extracting latent variables from the characteristic data; making a preliminary determination of whether each battery cell is defective based on the feature values ​​corresponding to the latent variables; and making a secondary determination of whether each battery cell is defective based on the preliminary determination result and the distribution of the feature values.

[0012] According to the embodiment, the step of making a secondary determination of whether each of the battery cells is defective may include the step of obtaining a distribution map showing the distribution of the feature values, and the step of making a secondary determination of whether each of the battery cells is defective based on the position of each of the battery cells on the distribution map.

[0013] According to the embodiment, the step of making a secondary determination of whether each battery cell is defective based on the position of each battery cell on the distribution diagram is characterized in that, among the cells that were determined to be normal as a result of the primary determination, cells whose distance from other cells that were determined to be normal on the distribution diagram is at or above a first level are made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be made to be at or above a second level among the cells that were determined to be defective as a result of the primary determination, cells whose density with other cells that were determined to be defective on the distribution diagram is at or above a second level are made to be

[0014] According to the embodiment, the step of extracting the latent variables is characterized by inputting the characteristic data into an autoencoder that has already been trained to extract the latent variables. [Effects of the Invention]

[0015] The battery diagnostic device and its operating method according to the embodiments disclosed herein can diagnose whether or not a battery cell is defective during the initial activation stage. This prevents defective battery cells from being released, reduces the risk of fire, and ensures customer trust.

[0016] The battery diagnostic device and its operation method according to the embodiments disclosed in this document can prevent fires caused by battery cell defects, prevent productivity decline due to fires, and enhance safety. In addition, various effects directly or indirectly grasped by this document can be provided.

Brief Description of Drawings

[0017] [Figure 1] It is a block diagram showing the configuration of a general battery pack. [[ID=**12]] [Figure 2] It is a block diagram showing a battery diagnostic device according to an embodiment disclosed in this document. [Figure 3] It is a diagram showing an example of the data flow of a battery diagnostic device according to an embodiment disclosed in this document. [Figure 4] It is a diagram showing an example of a primary judgment result and a distribution diagram according to an embodiment disclosed in this document. [Figure 5] It is a flowchart for explaining a battery diagnostic method according to an embodiment disclosed in this document. [Figure 6] It is a flowchart for explaining a secondary judgment process according to an embodiment disclosed in this document. [Figure 7] It is a block diagram showing the hardware configuration of a computing system for performing the operation method of a battery management device according to an embodiment disclosed in this document.

Modes for Carrying Out the Invention

[0018] Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include various modifications, equivalents, and / or alternatives of the embodiments of the present invention.

[0019] In this text, the singular form of a noun corresponding to an item may include one or more of the item unless the context clearly indicates otherwise. In this text, each phrase such as “A or B,” “at least one of A and B,” “at least one of A or B,” “A, B, or C,” “at least one of A, B, and C,” and “at least one of A, B, or C” may include any one of the items listed together in the phrase, or any possible combination thereof. Terms such as “first,” “second,” “first,” or “second” may be used merely to distinguish one component from other components and not to limit the component in any other respect (e.g., importance or order). When one (e.g., the first) component is referred to as being "connected" or "linked" to another (e.g., the second) component, with or without the terms "functionally" or "communically," this means that the first component may be connected to the other component directly (e.g., by wire), wirelessly, or via the third component.

[0020] Each component (e.g., module or program) described herein may include one or more individuals. According to various embodiments, one or more components or operations of the component may be omitted, or one or more other components or operations may be added. Alternatively or additionally, multiple components (e.g., modules or programs) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the multiple components identical or similar to those performed by the components of the multiple components before the integration. According to various embodiments, operations performed by modules, programs, or other components may be performed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be performed in a different order, omitted, or one or more other operations may be added.

[0021] As used herein, the term "module" or "... section" may include units implemented in hardware, software, or firmware, and may be used interchangeably with terms such as, for example, logic, logic blocks, components, or circuits. A module may be an integrated component, or may be the smallest unit or a part thereof of the components that perform one or more functions. For example, according to one embodiment, a module may be implemented in the form of an ASIC (application-specific integrated circuit).

[0022] Various embodiments of this document may be implemented as software (e.g., a program or an application) including one or more instruction words stored in a machine-readable storage medium (e.g., a memory). For example, a processor of a device may call at least one instruction of the one or more instruction words stored from the storage medium and execute it. This enables the device to be operated to perform at least one function according to the at least one called instruction word. The one or more instruction words may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory" only means that the storage medium is a tangible device and does not include a signal (e.g., an electromagnetic wave), and this term does not distinguish between cases where data is stored semi-permanently and temporarily in the storage medium.

[0023] FIG. 1 is a block diagram showing the configuration of a general battery pack. Referring to FIG. 1, a battery control system including a battery pack 1 according to an embodiment of the present invention and a host controller 2 included in a host system is schematically shown.

[0024] As shown in Figure 1, the battery pack 1 consists of one or more battery cells and includes a plurality of rechargeable battery cells 10, a switching unit 14 connected in series to the (+) terminal side or (-) terminal side of the plurality of battery cells 10 for controlling the flow of charge and discharge current to the plurality of battery cells 10, and a battery management system 20 that monitors the voltage, current, temperature, etc. of the battery pack 1 and controls and manages to prevent overcharging and over-discharging. In this case, the battery pack 1 can be provided with a plurality of battery cells 10, sensors 12, switching units 14, and battery management systems 20.

[0025] Here, the switching unit 14 is an element for controlling the flow of current for charging or discharging multiple battery cells 10, and for example, at least one relay, electromagnetic contactor, etc. can be used depending on the specifications of the battery pack 1.

[0026] The battery management system 20 is an interface that receives input values ​​of the various parameters described above, and may include multiple terminals and circuits connected to these terminals that process the input values. The battery management system 20 can also control the ON / OFF state of a switching unit 14, such as a relay or contactor, and can be connected to multiple battery cells 10 to monitor the state of each of the multiple battery cells 10. According to one embodiment, the battery management system 20 may include the battery diagnostic device 100 shown in Figure 2. According to another embodiment, the battery management system 20 may be a different system from the battery diagnostic device 100 shown in Figure 2. That is, the battery diagnostic device 100 shown in Figure 2 may be included in the battery pack 1, or it may be configured as another device outside the battery pack 1. Furthermore, the operation of the battery diagnostic device 100 described below may be performed by a BMS (Battery management system) in the vehicle, as well as by various devices such as a server, cloud, charger, or charger / discharger.

[0027] The higher-level controller 2 can transmit control signals to the battery management system 20 for multiple battery cells 10. This allows the battery management system 20 to be controlled based on the signals applied from the higher-level controller 2.

[0028] Figure 2 is a block diagram showing a battery diagnostic device according to one embodiment disclosed in this document. Referring to Figure 2, the battery diagnostic device 100 according to one embodiment disclosed in this document may include an information acquisition unit 110 and a controller 120. According to the embodiment, the battery diagnostic device 100 may be included in the battery management system 20 of Figure 1, or it may be a different device from the battery management system 20 of Figure 1.

[0029] The battery diagnostic device 100 can analyze the characteristic data of the battery cells contained in the battery pack 1 and determine whether each battery cell is defective. The battery diagnostic device 100 can analyze the characteristic data of each battery cell absolutely and relatively, thereby improving the accuracy of the battery cell defect judgment.

[0030] The information acquisition unit 110 can acquire characteristic data for each battery cell. The characteristic data for each battery cell can include various types of data related to the operation of the battery cell. For example, the characteristic data for a battery cell can include voltage data, current data, temperature data, capacity data, etc. According to the embodiment, the characteristic data for a battery cell can include differential capacity data (dQ / dV). Here, the capacity of the battery cell may mean the amount of charge of the battery cell. The information acquisition unit 110 can also acquire data such as SOH and SOC for each battery cell.

[0031] According to the embodiment, the information acquisition unit 110 can acquire characteristic data of the battery cell in the low-current section during the initial activation of the battery cell. If a fire occurs in a battery cell, it can damage surrounding cells, potentially reducing productivity and creating safety problems. To prevent this, the battery diagnostic device 100 can acquire battery data in the low-current section during the initial activation of the battery cell to diagnose defects in the battery cell. In other words, the battery diagnostic device 100 can quickly detect defects by diagnosing them in the initial activation stage, reduce the risk of fire by acquiring data in the low-current section, and enable defect detection.

[0032] For example, the battery diagnostic device 100 can diagnose battery defects during the pre-sale verification phase after the battery cells have been manufactured. As one example, the battery diagnostic device 100 can connect a low-current device to the manufactured battery cells, supply a low current to the battery cells to activate them, and acquire characteristic data of the battery cells in the low-current supply section to diagnose defects. With a low-current device, the detection accuracy of voltage, current, etc. is higher compared to a general charger / discharger, enabling more accurate defect detection.

[0033] Thus, the battery diagnostic device 100 can diagnose defects in battery cells at the initial activation stage, thereby preventing fire accidents. Furthermore, by detecting defects in battery cells at the initial activation stage, the battery diagnostic device 100 can prevent defective cells from being released or supplied, thereby ensuring customer trust, preventing fires, preventing productivity losses due to fires, and enhancing safety.

[0034] The controller 120 can extract latent variables from the characteristic data of the battery cell. The latent variables may implicitly include the characteristics of the battery cell's characteristic data. According to the embodiment, the controller 120 can input the characteristic data into an autoencoder that has already been trained to extract latent variables. The latent variables extracted by the autoencoder may be at least one.

[0035] An autoencoder can refer to a neural network that is trained to output data identical to the input data. An autoencoder can learn the process of extracting features from input data and reconstructing the input data from the extracted features. Through this learning process, an autoencoder can extract features (latent variables) that reflect the characteristics of the input data.

[0036] According to the embodiment, the autoencoder can learn the process of extracting latent variables upon input of characteristic data relating to normal cells. Data relating to normal cells can be stored in a database (not shown), memory (e.g., memory 1020 in Figure 7), etc., and can be used as training data in training the autoencoder. More specifically, the autoencoder can learn the process of extracting latent variables in the process of reconstructing the input characteristic data relating to normal cells.

[0037] The controller 120 can make a preliminary determination of whether each battery cell is defective based on the feature values ​​corresponding to the latent variables. The feature values ​​corresponding to the latent variables can mean values ​​that represent characteristic data in relation to the latent variables. For example, the feature values ​​can be represented as vectors with respect to the latent variables.

[0038] The controller 120 can make a preliminary determination of whether a battery cell is defective based on the fact that, since the autoencoder learns data about normal cells and extracts latent variables, when characteristic data about a defective cell is input to the already learned autoencoder, the feature values ​​of the latent variables of the defective cell show a different trend from the feature values ​​of the normal cell. For example, the controller 120 can set a reference value for the feature values ​​used to determine whether a battery cell is good or bad, and can determine whether each battery cell is defective by comparing the reference value with the feature values. In this case, the reference value can be set statistically or experimentally. The controller 120 can store the preliminary determination result of whether each battery cell is defective, and the preliminary determination result can include the defect determination result for each battery cell. In other words, the controller 120 can temporarily determine whether each battery cell is good or bad by the preliminary determination.

[0039] Conventional failure detection methods using autoencoders determine defects by using reconstruction errors, which represent the discrepancy between the autoencoder's input data and the reconstructed output data. However, the autoencoder according to the embodiment disclosed in this document does not utilize the discrepancy with the reconstructed data, but rather utilizes latent variables inherently extracted during the autoencoder's input data reconstruction process, thus representing a method entirely different from existing failure detection methods.

[0040] According to the embodiment, the controller 120 can input feature values ​​into a pre-trained model to generate a primary decision result. The pre-trained model can include learning models such as machine learning (ML), deep learning (DL), and reinforcement learning. The pre-trained model can include, but is not limited to, models such as CNNs (Convolutional Neural Networks) such as GoogleNet, AlexNet, and VGG Network, R-CNN (Region with Convolutional Neural Network), and FCNN (Fully Convolutional Neural Network).

[0041] A pre-trained model can learn the process of determining the quality of a battery cell from its feature values. In other words, a pre-trained model may take feature values ​​as input and output the result of the quality determination. For example, a pre-trained model may be a machine learning-based classification model.

[0042] The controller 120 can make a secondary determination of whether each battery cell is defective based on the primary determination result and the distribution of feature values. The secondary determination result may be the final determination regarding the defect status of each battery cell.

[0043] The controller 120 can make a secondary determination of whether a battery cell is defective by considering both the defect judgment result and the distribution of feature values ​​in the primary determination. The primary and secondary determination results for each battery cell do not have to be the same; for example, a battery cell that was judged normal in the primary determination may be judged defective in the secondary determination.

[0044] According to the embodiment, the controller 120 can obtain a distribution map showing the distribution of feature values. For example, the controller 120 can obtain a distribution map by showing the distribution of feature values ​​on a latent space where each axis is a latent variable. The controller 120 can also show the primary judgment result on the distribution map. That is, the controller 120 can represent normal cells and defective cells on the distribution map in a way that distinguishes them based on the result of the primary judgment. For example, the controller 120 can represent defective cells and normal cells on the distribution map using different colors, shapes, symbols, etc.

[0045] According to the embodiment, the controller 120 can make a secondary determination of whether each battery cell is defective based on its position on the distribution diagram. Since each battery cell has a feature value corresponding to characteristic data, the position of the feature value on the distribution diagram can be interpreted in the same way as the position of the battery cell. For example, even if a cell is judged to be normal as a result of the primary determination, the controller 120 can make a secondary determination of it as defective if its trend differs from that of a typical normal cell. As another example, even if a cell is judged to be defective as a result of the primary determination, the controller 120 can make a secondary determination of it as a normal cell if its trend is similar to that of a typical normal cell. In this way, the controller 120 can relatively determine whether a battery cell is defective by considering the position of each battery cell and its relationship to other battery cells.

[0046] According to the embodiment, the controller 120 can make a secondary determination of a cell as defective if, as a result of the primary determination, the distance from other cells that are determined to be normal on the distribution diagram is at or above the first level. The distance between cells may include the Euclidean distance on the distribution diagram. For example, the distance between a particular cell and other cells may be the average value of the Euclidean distances between that cell and other cells. The first level can be determined statistically or experimentally based on the center distance of the cells that are determined to be normal as a result of the primary determination. For example, the first level can be determined based on the average value of the center distances of the cells that are determined to be normal as a result of the primary determination.

[0047] In other words, the controller 120 can make a secondary judgment that a cell is defective, even if it is deemed normal based on the initial judgment, if it is located far from other cells that have been judged to be normal, because its characteristics differ from those of the other cells.

[0048] According to the embodiment, the controller 120 can make a secondary determination of cells that, as a result of the primary determination, are deemed defective, and whose density relative to other cells deemed defective on the distribution map is at or above the second level, as normal. The density can be determined based on the number of cells contained within a space of a specific size. The second level can be determined statistically or experimentally based on the density of cells on the distribution map. For example, the second level can be determined based on the average density of cells deemed normal and the average density of cells deemed defective as a result of the primary determination. As an example, the controller 120 can make a secondary determination of cells that, as a result of the primary determination, are deemed defective, and whose density is higher than the average density of cells deemed normal as a result of the primary determination, as normal.

[0049] In other words, if the density of cells judged to be defective is high, the controller 120 can make a secondary judgment that the cells are normal because their characteristics are similar and there is a high probability that they are actually normal and not defective.

[0050] If the secondary assessment confirms that a battery cell is defective, the controller 120 can provide the user with information about the defective battery cell. For example, the controller 120 can provide information about the defective battery cell to the user terminal via a communication unit (not shown), or it can provide information about the defective battery cell via a display installed in the vehicle or charger.

[0051] Figure 3 shows an example of the decision-making process of a battery diagnostic device according to one embodiment disclosed in this document. Referring to Figure 3, the battery diagnostic device 100 can extract latent variables from the characteristic data of the battery cells to make a preliminary determination of whether each battery cell is defective, and then make a secondary determination using the results of the preliminary determination.

[0052] First, the battery diagnostic device 100 can input characteristic data of the battery cells to the autoencoder 121. The autoencoder 121 can extract latent variables from the characteristic data. The latent variables extracted by the autoencoder 121 may be at least one.

[0053] The battery diagnostic device 100 can input the extracted latent variables into a pre-trained model 123. The pre-trained model 123 can generate a primary judgment result for each battery cell based on the feature values ​​corresponding to the latent variables. The pre-trained model can learn the process of determining whether a battery cell is defective from the feature values, and the primary judgment result can include whether each battery cell is defective or not.

[0054] The battery diagnostic device 100 can perform a secondary judgment using the primary judgment result. The battery diagnostic device 100 can input the primary judgment result and the feature value distribution map into the relative judgment model 125 to perform a secondary judgment on whether each battery cell is defective. The relative judgment model 125 may be a neural network model or a rule-based model. If the relative judgment model 125 is a neural network model, it may be a different model from the already trained model 123 that performs the primary judgment.

[0055] The autoencoder 121, the already trained model 123, and the relative decision model 125 may be part of the controller 120 or they may be physically separate external components.

[0056] Figure 4 shows an example of the primary judgment result and distribution map according to one embodiment disclosed in this document. Referring to Figure 4, the controller 120 can obtain a distribution diagram 400 showing the distribution of characteristic values ​​for each battery cell.

[0057] The controller 120 can make a preliminary judgment for each battery cell based on its characteristic values. The controller 120 can display the preliminary judgment results on the distribution diagram 400 so that they can be viewed. For example, as shown in Figure 4, the controller 120 can distinguish between defective cells and normal cells by showing defective cells 410 as circles and normal cells 420 as rectangles based on the results of the preliminary judgment.

[0058] In Figure 4, X1, X2, and X3, corresponding to each axis (x, y, z), may represent latent variables extracted by the autoencoder 121. As another example, the autoencoder 121 can extract four or more latent variables, and the controller 120 can select some of the extracted latent variables (X1, X2, X3) to obtain a distribution map.

[0059] Figure 5 is a flowchart illustrating a battery diagnostic method according to one embodiment disclosed in this document. Referring to Figure 5, the battery diagnostic method may include the steps of: acquiring characteristic data for each battery cell (S100); extracting latent variables from the characteristic data (S200); making a preliminary determination of whether each battery cell is defective based on the feature values ​​corresponding to the latent variables (S300); and making a secondary determination of whether each battery cell is defective based on the preliminary determination result and the distribution of feature values ​​(S400).

[0060] In step S100, the information acquisition unit 110 can acquire characteristic data for each battery cell. In this embodiment, the characteristic data for each battery cell may include differential capacitance data (dQ / dV).

[0061] In step S200, the controller 120 can extract latent variables from the characteristic data of the battery cell. In one embodiment, the controller 120 can input the characteristic data of the battery cell into an autoencoder to extract latent variables.

[0062] In step S300, the controller 120 can make a preliminary determination of whether each battery cell is defective based on its feature values. In this embodiment, the controller 120 can input the feature values ​​of each battery cell into a pre-learned model to obtain a preliminary determination result. That is, the controller 120 can make an absolute determination based on the feature values ​​for each battery cell.

[0063] In step S400, the controller 120 can make a secondary determination of whether each battery cell is defective based on the primary determination result and the distribution of feature values. In other words, the controller 120 can make a relative determination by considering the distribution of feature values ​​for each battery cell.

[0064] Figure 6 is a flowchart illustrating the secondary decision-making process according to one embodiment disclosed in this document. Referring to Figure 6, step S400 may include steps of obtaining a distribution map showing the distribution of feature values ​​(S210) and making a secondary determination of whether each battery cell is defective based on its position on the distribution map (S220).

[0065] In step S210, the controller 120 can obtain a distribution map showing the distribution of feature values. For example, the controller 120 can obtain a multidimensional latent space with latent variables as each axis, and obtain a distribution map by indicating the positions of feature values ​​on the latent space.

[0066] In step S220, the controller 120 can make a secondary determination of whether each battery cell is defective based on its position on the distribution diagram. In this embodiment, the controller 120 can make a secondary determination of defective cells among those that were determined to be normal as a result of the primary determination, if the distance from other cells that were determined to be normal on the distribution diagram is at or above the first level, and can make a secondary determination of normal cells among those that were determined to be defective as a result of the primary determination, if the density of other cells that were determined to be defective on the distribution diagram is at or above the second level.

[0067] Figure 7 is a block diagram showing the hardware configuration of a computing system for performing the operation method of a battery diagnostic device according to one embodiment disclosed in this document.

[0068] Referring to Figure 7, the computing system 1000 according to one embodiment disclosed in this document may include an MCU 1010, a memory 1020, an input / output interface 1030, and a communication interface 1040.

[0069] The MCU1010 may be a processor that executes various programs stored in memory 1020 (for example, a battery cell characteristic data acquisition program, a latent variable extraction program, a distribution map generation program, a battery cell diagnostic program, etc.), processes various information including battery cell characteristic data and latent variables through such programs, and performs the functions of the controller included in the battery diagnostic device shown in Figure 2 above.

[0070] Memory 1020 can store various programs, such as a battery cell characteristic data acquisition program, a latent variable extraction program, a distribution map generation program, and a battery cell diagnostic program. Memory 1020 can also store various information, including battery cell characteristic data and latent variables.

[0071] Multiple such memory 1020s may be provided as needed. Memory 1020 may be volatile memory or non-volatile memory. As volatile memory, RAM, DRAM, SRAM, etc., can be used for memory 1020. As non-volatile memory, ROM, PROM, EAROM, EPROM, EEPROM, flash memory, etc., can be used for memory 1020. The examples of memory 1020 listed above are merely illustrative and are not limiting.

[0072] The input / output interface 1030 can provide an interface that connects input devices (not shown), such as keyboards, mice, and touch panels, with output devices (not shown), such as displays, and the MCU 1010, enabling data transmission and reception.

[0073] The communication interface 1040 is configured to send and receive various data with a server and may be various devices that support wired or wireless communication. For example, a battery diagnostic device can send and receive various information, including the SOC, OCV, and parameters of battery cells, from a separately provided external server via the communication interface 1040.

[0074] Thus, the computer program according to one embodiment disclosed in this document may be recorded in memory 1020 and processed by MCU 1010 to be implemented as a module that performs, for example, the functions shown in Figure 2.

[0075] Although all components constituting the embodiments disclosed in this document have been described as operating either as a single unit or in combination, the embodiments disclosed in this document are not necessarily limited to such embodiments. That is, within the scope of the purpose of the embodiments disclosed in this document, all components may operate in combination of one or more units.

[0076] Furthermore, terms such as “includes,” “constitutes,” or “possesses,” as described above, mean that they may contain the component in question, and not exclude other components, unless otherwise specified. All terms, including technical or scientific terms, have the same meaning as those generally understood by a person of ordinary skill in the art to which the embodiments disclosed herein belong, unless otherwise specified. Commonly used terms, such as those defined in dictionaries, should be interpreted to be consistent with their meaning in the context of the relevant technology, and not to be interpreted in an ideal or overly formal sense unless explicitly defined herein.

[0077] The above description is merely illustrative of the technical concept disclosed herein, and any person with ordinary skill in the art to which the embodiments disclosed herein belong can make various modifications and variations without departing from the essential characteristics of the embodiments disclosed herein. Therefore, the embodiments disclosed herein are for illustrative purposes only, not to limit the technical concept of the embodiments disclosed herein, and the scope of the technical concept disclosed herein is not limited by such embodiments. The scope of protection of the technical concept disclosed herein must be interpreted according to the claims described below, and all technical concepts within an equivalent scope should be interpreted as being included in the scope of rights of this document.

Claims

1. An information acquisition unit that acquires characteristic data for each battery cell, Extract latent variables from the aforementioned characteristic data, Based on the characteristic values ​​corresponding to the aforementioned latent variables, a preliminary determination is made as to whether each of the battery cells is defective. A controller that makes a secondary determination of whether each of the battery cells is defective based on the results of the primary determination and the distribution of the feature values, Battery diagnostic device, including

2. The aforementioned controller, A distribution diagram showing the distribution of the aforementioned feature values ​​is obtained, The battery diagnostic device according to claim 1, which makes a secondary determination of whether each of the battery cells is defective based on the position of each of the battery cells on the distribution diagram.

3. The aforementioned controller, Based on the results of the initial assessment, among the battery cells that were judged to be normal, any battery cell whose distance from other battery cells judged to be normal on the distribution diagram is at or above the first level is judged to be defective in the second assessment. The battery diagnostic device according to claim 2, wherein, as a result of the initial judgment, among the battery cells that are judged to be defective, the battery cells that are densely packed with other battery cells that are judged to be defective on the distribution diagram at a density of 2 levels or higher are judged to be normal in a secondary judgment.

4. The aforementioned controller, The battery diagnostic device according to claim 1, wherein the characteristic data is input to an autoencoder that has already been trained to extract the latent variables.

5. The battery diagnostic device according to claim 4, wherein the autoencoder learns the process of extracting the latent variables by receiving characteristic data relating to a normal battery cell.

6. The aforementioned controller, The battery diagnostic device according to claim 1, wherein the aforementioned feature values ​​are input into a model that has already been learned to obtain the result of the initial judgment.

7. The battery diagnostic device according to claim 1, wherein the characteristic data includes differential capacitance data (dQ / dV) of the battery cell.

8. The aforementioned information acquisition unit, The battery diagnostic device according to claim 1, which acquires the characteristic data in the low current section during the initial activation of the battery cell.

9. Steps to acquire characteristic data for each battery cell, The steps include extracting latent variables from the aforementioned characteristic data, A step of making a preliminary determination of whether each of the battery cells is defective based on the characteristic values ​​corresponding to the latent variables, The steps include: making a secondary determination of whether each of the battery cells is defective based on the results of the primary determination and the distribution of the characteristic values; Battery diagnostic methods, including those mentioned above.

10. The step of making a secondary determination of whether each of the aforementioned battery cells is defective is as follows: The steps include obtaining a distribution map showing the distribution of the aforementioned feature values, The battery diagnostic method according to claim 9, comprising the step of making a secondary determination of whether each of the battery cells is defective based on the position of each of the battery cells on the distribution diagram.

11. The step of making a secondary determination of whether each of the battery cells is defective, based on the position of each of the battery cells on the distribution diagram, Based on the results of the initial assessment, among the battery cells that were judged to be normal, any battery cell whose distance from other battery cells judged to be normal on the distribution diagram is at or above the first level is judged to be defective in the second assessment. The battery diagnostic method according to claim 10, wherein, as a result of the initial judgment, among the battery cells that are judged to be defective, the battery cells that are densely packed with other battery cells that are judged to be defective on the distribution diagram at a level of 2 or higher are judged to be normal in a secondary judgment.

12. The step of extracting the aforementioned latent variables is: The battery diagnostic method according to claim 9, wherein the characteristic data is input to an autoencoder that has already been trained to extract the latent variables.

Images