Battery diagnostic device and operating method thereof

Abstract

According to an embodiment disclosed in the present document, a battery diagnostic device may comprise: an information-obtaining unit for obtaining characteristic data of each battery cell; and a controller that extracts a latent variable from the characteristic data, primarily determines whether each battery cell is defective on the basis of a feature value corresponding to the latent variable, and secondarily determines whether each battery cell is defective on the basis of the result of the primary determination and the distribution of the feature values.

Description

Battery diagnostic device and its operating method

[0001] Cross-citation with related applications

[0002] This invention claims the benefit of priority from Korean Patent Application No. 10-2023-0097064, filed July 25, 2023, the entire contents of which are incorporated herein by reference.

[0003] Technology field

[0004] The embodiments disclosed in this document relate to a battery diagnostic device and an operating method thereof.

[0005] Recently, research and development on secondary batteries has been actively conducted. Here, secondary batteries are rechargeable and include both conventional Ni / Cd and Ni / MH batteries, as well as recent lithium-ion batteries. Among secondary batteries, lithium-ion batteries have the advantage of having a much higher energy density than conventional Ni / Cd and Ni / MH batteries. Furthermore, lithium-ion batteries can be manufactured in a compact and lightweight form, making them suitable for use as power sources for mobile devices. Recently, their use has expanded to include power sources for electric vehicles, attracting attention as a next-generation energy storage medium.

[0006] Since there is a risk of fire if a battery cell contained in a battery pack has a defect such as a short circuit, exposure of the negative electrode, or lithium deposition, it is necessary to detect defective battery cells early and take action before a fire occurs.

[0007] One purpose of the embodiments disclosed in this document is to provide a battery diagnostic device and an operating method thereof that can prevent the risk of fire by detecting defects in a battery cell at an early stage.

[0008] 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 will be clearly understood by those skilled in the art from the descriptions below.

[0009] According to an embodiment disclosed in this document, a battery diagnosis device may include an information acquisition unit that acquires characteristic data of each battery cell, a controller that extracts latent variables from the characteristic data, determines whether each battery cell is defective based on a characteristic value corresponding to the latent variable, and determines whether each battery cell is defective based on a result of the first judgment and a distribution of the characteristic values.

[0010] According to an embodiment, the controller may obtain a distribution map representing the distribution of the feature values, and secondarily determine whether each battery cell is defective based on the location of each battery cell on the distribution map.

[0011] According to an embodiment, the controller may determine as a second time as defective a cell whose distance from other cells determined as normal on the distribution map among the cells determined as normal as a result of the first determination is greater than or equal to a first level, and may determine as normal a second time as defective a cell whose density from other cells determined as normal on the distribution map is greater than or equal to a second level, among the cells determined as defective as a result of the first determination.

[0012] According to an embodiment, the controller can input the characteristic data into a pre-trained autoencoder to extract the latent variable.

[0013] According to an embodiment, the autoencoder can learn a process of extracting the latent variable by inputting characteristic data for a normal cell.

[0014] According to an embodiment, the controller can obtain the first judgment result by inputting the feature value into a pre-learned model.

[0015] According to an embodiment, the characteristic data may include differential capacity data (dQ / dV) of the battery.

[0016] According to an embodiment, the information acquisition unit can acquire the characteristic data in a low current section at the initial stage of activation of the battery cell.

[0017] According to an embodiment disclosed in this document, a battery diagnosis method may include a step of acquiring characteristic data of each battery cell, a step of extracting a latent variable from the characteristic data, a step of first determining whether each battery cell is defective based on a characteristic value corresponding to the latent variable, and a step of second determining whether each battery cell is defective based on a result of the first determination and a distribution of the characteristic value.

[0018] According to an embodiment, the step of secondarily determining whether each of the battery cells is defective may include the step of obtaining a distribution diagram representing the distribution of the feature values ​​and the step of secondarily determining whether each of the battery cells is defective based on the location of each of the battery cells on the distribution diagram.

[0019] According to an embodiment, the step of secondarily determining whether each of the battery cells is defective based on the location of each of the battery cells on the distribution map may be characterized in that, among the cells determined to be normal as a result of the first determination, a cell whose distance from other cells determined to be normal on the distribution map is at least a first level is secondarily determined as defective, and among the cells determined to be defective as a result of the first determination, a cell whose density from other cells determined to be defective on the distribution map is at least a second level is secondarily determined as normal.

[0020] According to an embodiment, the step of extracting the latent variable may be characterized by inputting the characteristic data into a pre-trained autoencoder to extract the latent variable.

[0021] The battery diagnostic device and its operating method according to the embodiments disclosed in this document can diagnose whether a battery cell is defective during the initial activation stage. Accordingly, the leakage of defective battery cells can be prevented, reducing the risk of fire and securing customer trust.

[0022] The battery diagnostic device and its operating method according to the embodiments disclosed in this document can prevent fire caused by defects in battery cells, thereby preventing a decrease in productivity due to fire and improving safety.

[0023] In addition, various effects may be provided, either directly or indirectly, through this document.

[0024] Figure 1 is a block diagram showing the configuration of a typical battery pack.

[0025] FIG. 2 is a block diagram showing a battery diagnostic device according to one embodiment disclosed in this document.

[0026] FIG. 3 is a diagram showing an example of a data flow of a battery diagnostic device according to one embodiment disclosed in this document.

[0027] FIG. 4 is a diagram showing an example of a first judgment result and distribution according to one embodiment disclosed in this document.

[0028] FIG. 5 is a flowchart illustrating a battery diagnosis method according to one embodiment disclosed in this document.

[0029] Figure 6 is a flowchart illustrating a secondary judgment process according to one embodiment disclosed in this document.

[0030] FIG. 7 is a block diagram showing the hardware configuration of a computing system for performing an operating method of a battery management device according to one embodiment disclosed in this document.

[0031] Hereinafter, various embodiments of the present invention will be described with reference to the attached drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that the present invention encompasses various modifications, equivalents, and / or alternatives of the embodiments.

[0032] In this document, the singular form of a noun corresponding to an item may include one or more of said items, unless the context clearly indicates otherwise. In this document, phrases 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 each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish the corresponding element from other corresponding elements, and do not limit the corresponding elements in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as being “coupled” or “connected” to another component (e.g., a second component), with or without the terms “functionally” or “communicatively,” it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0033] Each component (e.g., a module or a program) described in this document may include one or more entities. According to various embodiments, one or more components or operations of the components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0034] The term "module" or "part" used in this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0035] Various embodiments of the present document may be implemented as software (e.g., a program or an application) including one or more instructions stored in a machine-readable storage medium (e.g., memory). For example, a processor of the device may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the device to operate to perform at least one function according to the at least one instruction called. The one or more instructions 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 contain a signal (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0036] Figure 1 is a block diagram showing the configuration of a typical battery pack.

[0037] Referring to FIG. 1, a battery control system including a battery pack (1) and an upper controller (2) included in an upper system according to one embodiment of the present invention is schematically illustrated.

[0038] As illustrated in FIG. 1, a battery pack (1) is composed of one or more battery cells and includes a plurality of battery cells (10) that are rechargeable and dischargeable, a switching unit (14) that is connected in series to the (+) terminal side or the (-) terminal side of the plurality of battery cells (10) to control the flow of charge and discharge current of 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 overdischarging, etc. In this case, the battery pack (1) may be equipped with a plurality of battery cells (10), sensors (12), switching units (14), and battery management systems (20).

[0039] Here, the switching unit (14) is a device for controlling the current flow for charging or discharging of a plurality of battery cells (10), and for example, at least one relay, magnetic contactor, etc. may be used depending on the specifications of the battery pack (1).

[0040] The battery management system (20) is an interface that receives values ​​measured from the various parameters described above, and may include a plurality of terminals and a circuit that is connected to the terminals and processes the values ​​received. In addition, the battery management system (20) may control the ON / OFF of a switching unit (14), for example, a relay or a contactor, and may be connected to a plurality of battery cells (10) and monitor the status of each of the plurality of battery cells (10). According to an embodiment, the battery management system (20) may include the battery management device (100) of FIG. 2. According to another embodiment, the battery management system (20) may be a different system from the battery management device (100) of FIG. 2. That is, the battery management device (100) of FIG. 2 may be included in the battery pack (1) or may be configured as another device external to the battery pack (1). In addition, the operation of the battery management device (100) below can be performed by a BMS (Battery Management System) in a vehicle, as well as by various devices such as a server, cloud, charger, or charger / discharger.

[0041] The upper controller (2) can transmit control signals for multiple battery cells (10) to the battery management system (20). Accordingly, the battery management system (20) can be controlled in operation based on the signals received from the upper controller (2).

[0042] FIG. 2 is a block diagram showing a battery diagnostic device according to one embodiment disclosed in this document.

[0043] Referring to FIG. 2, a battery diagnostic device (100) according to one embodiment disclosed in the present document may include an information acquisition unit (110) and a controller (120). Depending on the embodiment, the battery diagnostic device (100) may be included in the battery management system (20) of FIG. 1, or may be another device different from the battery management system (20) of FIG. 1.

[0044] A battery diagnostic device (100) can analyze characteristic data of battery cells included in a battery pack (1) to determine whether each battery cell is defective. The battery diagnostic device (100) can analyze the characteristic data of each battery cell absolutely and relatively to increase the accuracy of determining whether a battery cell is defective.

[0045] The information acquisition unit (110) can acquire characteristic data of each battery cell. The characteristic data of the battery cell may include various types of data related to the operation of the battery cell. For example, the characteristic data of the battery cell may include voltage data, current data, temperature data, capacity data, etc. According to an embodiment, the characteristic data of the battery cell may include differential capacity data (dQ / dV). Here, the capacity of the battery cell may refer to the amount of charge of the battery cell. In addition, the information acquisition unit (110) may further acquire data such as SOH and SOC of each battery cell.

[0046] According to an embodiment, the information acquisition unit (110) can acquire characteristic data of a battery cell in a low-current section during the initial activation of the battery cell. If a fire occurs in a battery cell, surrounding cells of the cell where the fire occurred may be damaged, which not only reduces productivity but may also cause safety issues. To prevent this, the battery diagnosis device (100) can diagnose defects in the battery cell by acquiring battery data during the low-current section during the initial activation of the battery cell. That is, the battery diagnosis device (100) can quickly detect defects through diagnosis during the initial activation, and can detect defects while reducing the risk of fire through data acquisition during the low-current section.

[0047] For example, the battery diagnostic device (100) can perform battery defect diagnosis during the pre-sale verification stage after the battery cell has been manufactured. For example, the battery diagnostic device (100) can perform defect diagnosis by connecting a low-current device to a manufactured battery cell, activating the battery cell by supplying a low current to the battery cell, and obtaining characteristic data of the battery cell during the low-current supply section. In the case of the low-current device, the detection accuracy of voltage, current, etc. is higher than that of a general charger / discharger, so that more accurate defect determination can be made.

[0048] In this way, the battery diagnostic device (100) can diagnose battery cell defects in the initial activation stage of the battery cell, thereby preventing fire accidents in advance. Furthermore, the battery diagnostic device (100) can secure customer trust by detecting battery cell defects in the initial activation stage of the battery cell, thereby preventing the leakage and supply of defective cells. Furthermore, by preventing fire occurrence, it can prevent productivity declines due to fire and enhance safety.

[0049] The controller (120) can extract latent variables from the characteristic data of the battery cell. The latent variables may be variables that implicitly contain the characteristics of the characteristic data of the battery cell. According to an embodiment, the controller (120) can input the characteristic data into a pre-trained auto-encoder to extract latent variables. The number of latent variables extracted by the auto-encoder may be at least one.

[0050] An autoencoder can refer to a neural network trained to output data identical to the input data. An autoencoder can learn to extract features from input data and then reconstruct the input data from these extracted features. Through the learning process, an autoencoder can extract features (latent variables) that reflect the characteristics of the input data.

[0051] According to an embodiment, an autoencoder can learn the process of extracting latent variables by receiving characteristic data for normal cells as input. The data for normal cells may be stored in a database (not shown), memory (e.g., memory (1020) of FIG. 7), etc., and may be utilized as training data in the autoencoder's training. More specifically, the autoencoder can learn the process of extracting latent variables by restoring characteristic data for input normal cells.

[0052] The controller (120) can initially determine whether each battery cell is defective based on a feature value corresponding to a latent variable. The feature value corresponding to a latent variable may refer to a value that expresses characteristic data in relation to the latent variable. For example, the feature value may be expressed as a vector of latent variables.

[0053] The controller (120) can make a primary judgment on whether a battery cell is defective based on the fact that, since the auto-encoder learns data on normal cells and extracts latent variables, when characteristic data on a defective cell is input to the pre-learned auto-encoder, the characteristic values ​​of the latent variables of the defective cell will exhibit a different tendency from the characteristic values ​​of the normal cell. For example, the controller (120) can set a reference value of the characteristic values ​​for judging whether a battery cell is defective, and can determine whether each battery cell is defective by comparing the reference value and the characteristic value. At this time, the reference value can be set statistically or experimentally. The controller (120) can store the primary judgment result on whether each battery cell is defective, and the primary judgment result can include the result of judging whether each battery cell is defective. That is, the controller (120) can temporarily determine whether each battery cell is defective through the primary judgment.

[0054] Conventional defect detection methods utilizing autoencoders use reconstruction errors, which represent the discrepancy between the autoencoder's input data and the restored output data, to determine defects. However, the autoencoder according to the embodiments disclosed in this document differs significantly from existing defect detection methods in that, rather than utilizing the discrepancy between the restored data and the restored data, it utilizes latent variables inherently extracted during the autoencoder's input data restoration process.

[0055] According to an embodiment, the controller (120) can input feature values ​​into a pre-learned model to generate a first judgment result. The pre-learned model may include a learning model such as machine learning (ML), deep learning (DL), reinforcement learning, etc. The pre-learned model may include a model such as a CNN (Convolutional Neural Network) such as GoogleNet, AlexNet, VGG Network, R-CNN (Region with Convolutional Neural Network), FCNN (Fully Convolutional Neural Network), etc., but is not limited to the above-mentioned models.

[0056] A pre-trained model can learn the process of determining battery cell quality from feature values. In other words, a pre-trained model can be a model that takes feature values ​​as input and outputs the quality / failure judgment result. For example, a pre-trained model could be a machine learning-based classifier model.

[0057] The controller (120) can make a secondary judgment on whether each battery cell is defective based on the primary judgment result and the distribution of characteristic values. The secondary judgment result may be the final judgment on whether each battery cell is defective.

[0058] The controller (120) can make a secondary judgment on whether a battery cell is defective by considering both the defective judgment result of the battery cell in the primary judgment and the distribution of the characteristic values. The primary and secondary judgment results of each battery cell may not be identical, and for example, a battery cell that is determined to be normal in the primary judgment may be determined to be defective in the secondary judgment.

[0059] According to an embodiment, the controller (120) can obtain a distribution diagram representing the distribution of feature values. For example, the controller (120) can obtain the distribution diagram by displaying the distribution of feature values ​​on a latent space in which each axis is a latent variable. In addition, the controller (120) can display the first judgment result together on the distribution diagram. That is, the controller (120) can express the first judgment result on the distribution diagram so that normal cells and defective cells can be distinguished. For example, the controller (120) can express defective cells and normal cells on the distribution diagram using different colors, shapes, symbols, etc.

[0060] According to an embodiment, the controller (120) can secondarily determine whether each battery cell is defective based on the location of each battery cell on the distribution map. Since each battery cell has a characteristic value corresponding to characteristic data, the location of the characteristic value on the distribution map can be interpreted in the same way as the location of the battery cell. For example, even if a cell is determined to be normal as a result of the first determination, the controller (120) can secondarily determine that it is defective if its tendency is different from that of general normal cells. In another example, even if the cells are determined to be defective as a result of the first determination, the controller (120) can determine that they are normal cells if their tendencies are similar. In this way, the controller (120) can relatively determine whether a battery cell is defective by considering the location of each battery cell and its relationship with other battery cells.

[0061] According to an embodiment, the controller (120) may secondarily determine as defective a cell, among cells determined as normal in the first judgment, if the distance between the cells and other cells determined as normal in the distribution map is greater than or equal to a first level. The distance between cells may include a Euclidean distance in the distribution map. For example, the distance between a specific cell and other cells may be the average value of the Euclidean distances between the cell and other cells. The first level may be determined statistically or experimentally based on the center distances of cells determined as normal in the first judgment. For example, the first level may be determined based on the average value of the center distances of cells determined as normal in the first judgment.

[0062] That is, even if a cell is judged normal in the first judgment, if it is different from other cells judged normal, the controller (120) can secondarily judge the cell as defective, predicting that the cell has different characteristics from other cells and is more likely to be defective.

[0063] According to an embodiment, the controller (120) may secondarily determine as normal cells, among cells determined as defective in the first judgment, cells whose density relative to other cells determined as defective in the distribution map is higher than a second level. The density may be determined based on the number of cells contained in a space of a specific size, etc. In addition, the second level may be determined statistically or experimentally based on the density of cells in the distribution map. For example, the second level may be determined based on the average density of cells determined as normal in the first judgment and the average density of cells determined as defective. For example, the controller (120) may secondarily determine as normal cells, among cells determined as defective in the first judgment, cells whose density is higher than the average density of cells determined as normal in the first judgment.

[0064] That is, if the density of cells judged as defective is high, the controller (120) can determine the cells as normal for a second time because the characteristics of the cells are similar, and thus the probability that the cells are normal is higher than the actual defective cells.

[0065] In addition, if the secondary judgment results confirm that the battery cell is defective, the controller (120) can provide information about the defective battery cell to the user. For example, the controller (120) can provide information about the defective battery cell to the user terminal via a communication unit (not shown), and can also provide information about the defective battery cell via a display provided in a vehicle or charger.

[0066] FIG. 3 is a diagram showing an example of a judgment process of a battery diagnostic device according to one embodiment disclosed in this document.

[0067] Referring to FIG. 3, the battery diagnostic device (100) can extract latent variables from characteristic data of battery cells to make a primary judgment on whether each battery cell is defective, and perform a secondary judgment using the primary judgment result.

[0068] First, the battery diagnostic device (100) can input characteristic data of a battery cell into an auto encoder (121). The auto encoder (121) can extract latent variables from the characteristic data. There can be at least one latent variable extracted from the auto encoder (121).

[0069] 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 judging whether a battery cell is defective from the feature values, and the primary judgment result can include whether each battery cell is defective.

[0070] The battery diagnosis device (100) can perform a secondary judgment using the first judgment result. The battery diagnosis device (100) can input the first judgment result and the distribution of the feature value into a 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, the relative judgment model (125) may be a different model from the pre-trained model (123) that performs the first judgment.

[0071] The autoencoder (121), the learned model (123) and the relative judgment model (125) may be part of the controller (120) or may be physically separate external components.

[0072] FIG. 4 is a diagram showing an example of a first judgment result and distribution according to one embodiment disclosed in this document.

[0073] Referring to FIG. 4, the controller (120) can obtain a distribution diagram (400) representing the distribution of characteristic values ​​of each battery cell.

[0074] The controller (120) can perform a first judgment on each battery cell based on the characteristic value. The controller (120) can express the first judgment result so that it can be confirmed on the distribution map (400). For example, as shown in FIG. 4, the controller (120) can distinguish between defective cells and normal cells by displaying defective cells (410) as circles based on the first judgment result and normal cells (420) as squares based on the first judgment result.

[0075] In Fig. 4, X1, X2, and X3 corresponding to each axis (x, y, z) may mean latent variables extracted by the auto encoder (121). In another example, the auto encoder (121) may extract four or more latent variables, and the controller (120) may select some of the extracted latent variables (X1, X2, X3) to obtain a distribution map.

[0076] FIG. 5 is a flowchart illustrating a battery diagnosis method according to one embodiment disclosed in this document.

[0077] Referring to FIG. 5, the battery diagnosis method may include a step of obtaining characteristic data of each battery cell (S100), a step of extracting latent variables from the characteristic data (S200), a step of first determining whether each battery cell is defective based on a characteristic value corresponding to the latent variable (S300), and a step of second determining whether each battery cell is defective based on a result of the first determination and a distribution of the characteristic values ​​(S400).

[0078] In step S100, the information acquisition unit (110) can acquire characteristic data of each battery cell. In an embodiment, the characteristic data of the battery cell can include differential capacity data (dQ / dV).

[0079] At step S200, the controller (120) can extract latent variables from characteristic data of the battery cell. In an embodiment, the controller (120) can input characteristic data of the battery cell into an autoencoder to extract latent variables.

[0080] At step S300, the controller (120) can make a preliminary judgment on whether each battery cell is defective based on the characteristic values. In an embodiment, the controller (120) can input the characteristic values ​​of each battery cell into a pre-trained model to obtain the preliminary judgment result. In other words, the controller (120) can perform an absolute judgment based on the characteristic values ​​for each battery cell.

[0081] At step S400, the controller (120) can make a secondary judgment on whether each battery cell is defective based on the primary judgment result and the distribution of the characteristic values. That is, the controller (120) can perform a relative judgment by considering the distribution of the characteristic values ​​of each battery cell.

[0082] Figure 6 is a flowchart illustrating a secondary judgment process according to one embodiment disclosed in this document.

[0083] Referring to FIG. 6, step S200 may include a step of obtaining a distribution diagram representing the distribution of feature values ​​(S210) and a step of secondarily determining whether each battery cell is defective based on the location of each battery cell on the distribution diagram (S220).

[0084] In step S210, the controller (120) can obtain a distribution diagram representing 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 diagram by indicating the location of feature values ​​on the latent space.

[0085] In step S220, the controller (120) may determine whether or not each battery cell is defective based on its location on the distribution map. In an embodiment, the controller (120) may determine as defective a cell among cells determined as normal in the first determination result, whose distance from other cells determined as normal in the distribution map is greater than or equal to a first level, and may determine as normal a cell among cells determined as defective in the first determination result, whose density from other cells determined as defective in the distribution map is greater than or equal to a second level.

[0086] FIG. 7 is a block diagram showing the hardware configuration of a computing system for performing an operating method of a battery management device according to one embodiment disclosed in this document.

[0087] Referring to FIG. 7, a computing system (1000) according to one embodiment disclosed in the present document may include an MCU (1010), a memory (1020), an input / output I / F (1030), and a communication I / F (1040).

[0088] The MCU (1010) may be a processor that executes various programs stored in the memory (1020) (e.g., a battery cell characteristic data collection program, a latent variable extraction program, a distribution generation program, a battery cell diagnosis program, etc.), processes various information including battery cell characteristic data and latent variables through these programs, and performs the functions of the controller included in the battery diagnosis device shown in FIG. 2 described above.

[0089] The memory (1020) can store various programs, such as a battery cell characteristic data collection program, a latent variable extraction program, a distribution map generation program, and a battery cell diagnosis program. In addition, the memory (1020) can store various information, including battery cell characteristic data, latent variables, and the like.

[0090] Such memories (1020) may be provided in multiple numbers as needed. The memories (1020) may be volatile memories or non-volatile memories. As volatile memories (1020), RAM, DRAM, SRAM, etc. may be used. As non-volatile memories (1020), ROM, PROM, EAROM, EPROM, EEPROM, flash memories, etc. may be used. The examples of the memories (1020) listed above are merely examples and are not limited to these examples.

[0091] The input / output I / F (1030) can provide an interface that enables data transmission and reception between an input device (not shown) such as a keyboard, mouse, or touch panel, and an output device (not shown) such as a display and the MCU (1010).

[0092] The communication I / F (1040) is a component capable of transmitting and receiving various data with the server, and may be any device capable of supporting wired or wireless communication. For example, the battery management device can transmit and receive various information, including the SOC, OCV, and parameters of the battery cell, from a separately provided external server via the communication I / F (1040).

[0093] In this way, a computer program according to one embodiment disclosed in this document may be implemented as a module that is recorded in a memory (1020) and processed by an MCU (1010) to perform each function illustrated in FIG. 2, for example.

[0094] Although all components constituting the embodiments disclosed in this document have been described as being combined or operating in combination as one, 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 of the components may be selectively combined and operated one or more times.

[0095] In addition, terms such as "include," "comprise," or "have" described above, unless specifically stated otherwise, mean that the corresponding component can be included, and therefore should be interpreted to include other components rather than excluding other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments disclosed in this document belong, unless otherwise defined. Commonly used terms, such as terms defined in a dictionary, should be interpreted to be consistent with the contextual meaning of the relevant technology, and shall not be interpreted in an idealized or overly formal sense, unless explicitly defined in this document.

[0096] The above description is merely an example of the technical idea disclosed in this document, and those skilled in the art to which the embodiments disclosed in this document pertain may make various modifications and variations without departing from the essential characteristics of the embodiments disclosed in this document. Therefore, the embodiments disclosed in this document are not intended to limit the technical idea of ​​the embodiments disclosed in this document, but to explain it, and the scope of the technical idea disclosed in this document is not limited by these embodiments. The scope of protection of the technical idea disclosed in this document should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of rights of this document.

Claims

1. An information acquisition unit that acquires characteristic data of each battery cell; and Extract latent variables from the above characteristic data, Based on the feature values ​​corresponding to the latent variables, the first determination is made as to whether each battery cell is defective. A battery diagnostic device including a controller that makes a second judgment on whether each battery cell is defective based on the first judgment result and the distribution of the feature value.

2. In paragraph 1, The above controller, Obtain a distribution diagram representing the distribution of the above feature values, A battery diagnostic device that makes a secondary judgment on whether each battery cell is defective based on the location of each battery cell on the distribution map.

3. In paragraph 2, The above controller, Among the cells judged as normal in the first judgment result, cells whose distance from other cells judged as normal in the distribution map is greater than the first level are judged as defective in the second judgment. A battery diagnostic device that determines, as a second step, as normal cells, cells that are determined to be defective in the first judgment result and have a density of at least a second level with other cells determined to be defective in the distribution diagram.

4. In paragraph 1, The above controller, A battery diagnostic device that inputs the above characteristic data into a pre-trained autoencoder to extract the above latent variable.

5. In paragraph 4, The above autoencoder is a battery diagnostic device that learns the process of extracting the latent variable by inputting characteristic data for a normal cell.

6. In paragraph 1, The above controller, A battery diagnosis device that inputs the above feature values ​​into a pre-learned model to obtain the first judgment result.

7. In paragraph 1, A battery diagnostic device, wherein the above characteristic data includes differential capacity data (dQ / dV) of the battery.

8. In paragraph 1, The above information acquisition unit, A battery diagnostic device that acquires the characteristic data in the low current section at the initial stage of activation of the battery cell.

9. Step of acquiring characteristic data of each battery cell; A step of extracting latent variables from the above characteristic data; A step of first determining whether each battery cell is defective based on the feature value corresponding to the latent variable; and A battery diagnosis method, comprising a step of secondarily determining whether each battery cell is defective based on the first judgment result and the distribution of the feature value.

10. In paragraph 9, The second step of judging whether each of the above battery cells is defective is as follows: A step of obtaining a distribution diagram representing the distribution of the above feature values; and A battery diagnosis method, comprising a step of secondarily determining whether each battery cell is defective based on the location of each battery cell on the distribution map.

11. In paragraph 10, The step of secondarily determining whether each battery cell is defective based on the location of each battery cell on the distribution map is as follows: Among the cells judged as normal in the first judgment result, cells whose distance from other cells judged as normal in the distribution map is greater than the first level are judged as defective in the second judgment. A battery diagnosis method characterized in that, among the cells judged as defective in the first judgment result, cells whose density with other cells judged as defective in the distribution map is at least a second level are judged as normal.

12. In paragraph 9, The step of extracting the above latent variables is: A battery diagnosis method characterized in that the latent variable is extracted by inputting the above characteristic data into a pre-trained autoencoder.

Images