Battery diagnosis device and battery diagnosis method

Abstract

According to embodiments of the present document, a battery diagnostic device includes a memory configured to store at least one instruction; and at least one processor configured to identify a first input voltage curve including a voltage curve measured by supplying a current having a specified waveform to a battery cell, input the first input voltage curve into a first model, the first model encodes the first input voltage curve and then recovers, and diagnose a state of the battery cell based on a feature vector identified according to the first model.

Description

Technical Field

[0001] Cross-references to related applications

[0002] This application claims priority to Korean Patent Application No. 10-2024-0087431, filed on July 3, 2024, and Korean Patent Application No. 10-2025-0078638, filed on June 16, 2025, the disclosures of which are incorporated herein by reference. Technical Field

[0004] The embodiments disclosed in this document relate to a battery diagnostic device and a battery diagnostic method. Background Technology

[0005] Recently, research and development on rechargeable batteries have been actively pursued. Here, a rechargeable battery is a battery that can be charged and discharged, and can be interpreted as including conventional Ni / Cd batteries, Ni / MH batteries, and more recently, lithium-ion batteries. The applications of rechargeable batteries have recently expanded to include power sources for electric vehicles, and they are attracting attention as a next-generation energy storage medium.

[0006] With the proliferation of electronic devices resulting from the Fourth Industrial Revolution, battery usage is increasing dramatically. Batteries are receiving significant attention as a fundamental energy source in various applications, and consequently, the importance of technologies for diagnosing battery status to improve battery performance and reliability is growing.

[0007] With recent advancements in artificial intelligence models, battery diagnostic devices utilizing battery data are being developed. Furthermore, multiple trained models can be used as tools to diagnose the condition of individual battery cells. Summary of the Invention

[0008] Technical issues

[0009] The embodiments disclosed in this document provide a battery diagnostic device and a battery diagnostic method that reduce the noise rate by using dimensionality-reduced feature vectors identified in a first model that performs encoding and then recovers the input data.

[0010] The embodiments disclosed in this document provide a battery diagnostic device and a battery diagnostic method that reduce the overfitting rate in a second model, which serves as a regression model, by reducing the noise occurrence rate.

[0011] The embodiments disclosed in this document provide a battery diagnostic device and a battery diagnostic method that improve the diagnostic accuracy of individual battery cell states based on feature vectors extracted by a first model that has undergone a pattern learning process.

[0012] The embodiments disclosed in this document provide a battery diagnostic device and a battery diagnostic method that reduce the decrease in training stability caused by high-dimensional data processing.

[0013] The embodiments disclosed in this document provide a battery diagnostic device and a battery diagnostic method that improve training efficiency even when the amount of input data is less than or equal to a specified amount.

[0014] The technical problems of the embodiments disclosed in this document are not limited to the above-described technical problems, and other technical problems not mentioned can be clearly understood by those skilled in the art from the following description.

[0015] Technical solution

[0016] The battery diagnostic device according to an embodiment of this document includes a memory configured to store at least one instruction and at least one processor configured to execute at least one instruction.

[0017] According to an embodiment, at least one processor may be configured to identify a first input voltage curve including a voltage curve measured by supplying a current having a specified waveform to a battery cell, inputting the first input voltage curve into a first model, encoding and then recovering the first input voltage curve in the first model, and diagnosing the state of the battery cell based on a feature vector identified according to the first model.

[0018] According to an embodiment, at least one processor can be configured to identify feature vectors based on latent vectors that are the processing results in a bottleneck layer included in a first model.

[0019] According to an embodiment, at least one processor may be configured to diagnose the state of a battery cell based on at least one of at least a portion of a feature vector, at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve based on a first model recovery, at least a portion of the derivative of the first input voltage curve, or any combination thereof.

[0020] According to an embodiment, at least one processor may be configured to diagnose the state of a battery cell based on at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, and at least a portion of the feature vector, based on the fact that all components of the feature vector fall within a pre-specified range; or to diagnose the state of a battery cell based on at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, based on the fact that at least one component of the feature vector does not fall within a pre-specified range.

[0021] According to an embodiment, at least one processor can be configured to input at least a portion of a feature vector into a second model, the second model outputting the state of a single battery cell.

[0022] According to an embodiment, at least one processor can be configured to supply current to a battery cell during a specified time period.

[0023] According to an embodiment, the state of a single battery cell may include a state of health (SOH).

[0024] According to an embodiment, the first model may include a Long Short-Term Memory Autoencoder (LSTM AE) model.

[0025] A battery diagnostic method according to another embodiment of this document includes: an operation of identifying a first input voltage curve, the first input voltage curve including a voltage curve measured by supplying a current having a specified waveform to a battery cell; an operation of inputting the first input voltage curve into a first model, the first model encoding and then recovering the first input voltage curve; and an operation of diagnosing the state of a battery cell based on a feature vector identified according to the first model.

[0026] According to an embodiment, the operation of diagnosing the state of a battery cell based on a feature vector identified according to a first model may include the operation of identifying a feature vector based on a latent vector that is a processing result in a bottleneck layer included in the first model.

[0027] According to an embodiment, the operation of diagnosing the state of a battery cell based on a feature vector identified according to a first model may include diagnosing the state of a battery cell based on at least a portion of the feature vector, at least a portion of a first input voltage curve, at least a portion of a recovered voltage curve recovered based on the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof.

[0028] According to an embodiment, the operation of identifying feature vectors based on the latent vectors of the processing results in the bottleneck layer included in the first model may include diagnosing the state of a battery cell based on at least a portion of the first input voltage curve, at least a portion of the recovery voltage curve recovered by the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, and at least a portion of the feature vector, based on the fact that at least one component of the feature vector does not fall within the pre-specified range.

[0029] According to an embodiment, the operation of diagnosing the state of a battery cell based on a feature vector identified by a first model may include an operation of inputting at least a portion of the feature vector into a second model, wherein the second model outputs the state of the battery cell.

[0030] According to an embodiment, the operation of identifying a first input voltage curve, which includes a voltage curve measured by supplying a current having a specified waveform to a battery cell, may include the operation of supplying current to the battery cell within a specified time period.

[0031] According to an embodiment, the state of a single battery cell may include a state of health (SOH).

[0032] According to an embodiment, the first model may include a Long Short-Term Memory Autoencoder (LSTM AE) model.

[0033] According to yet another embodiment of this document, a computer-readable recording medium records a program for performing a battery diagnostic method on a computer.

[0034] A battery diagnostic device according to another embodiment of this document includes a memory for storing at least one instruction and at least one processor for executing at least one instruction.

[0035] According to an embodiment, at least one processor is configured to identify a first input voltage curve by supplying a current with a first waveform to a battery cell, or to identify a second input voltage curve based on a current with a second waveform output from a battery cell, input the first input voltage curve or the second input voltage curve into a first model, which encodes and then recovers the first input voltage curve or the second input voltage curve, and diagnoses the state of the battery cell based on the feature vector identified according to the first model.

[0036] According to an embodiment, at least one processor can be configured to identify a first input voltage curve when the charge level of a battery cell falls within a pre-specified state range, and to identify a second input voltage curve when the charge level of a battery cell does not fall within the pre-specified state range.

[0037] According to an embodiment, the pre-specified state range may include values ​​smaller than the pre-specified state value.

[0038] According to an embodiment, at least one processor may be configured to identify a feature vector based on at least a portion of the components of a potential vector as a result of processing in a bottleneck layer included in a first model.

[0039] According to an embodiment, at least one processor may be configured to diagnose the state of a battery cell based on at least a portion of a feature vector, at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, or to diagnose the state of a battery cell based on at least a portion of a feature vector, at least a portion of a second input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the second input voltage curve, or any combination thereof.

[0040] According to an embodiment, at least one processor can be configured to input at least a portion of the feature vector into a second model of the state of the output battery cell.

[0041] According to an embodiment, the first model may include a Long Short-Term Memory Autoencoder (LSTM AE) model.

[0042] A battery diagnostic method according to another embodiment of this document includes: identifying a first input voltage curve by supplying a current having a first waveform to a battery cell or identifying a second input voltage curve based on a current having a second waveform output from the battery cell; inputting the first input voltage curve or the second input voltage curve into a first model, the first model encoding and then recovering the first input voltage curve or the second input voltage curve; and diagnosing the state of the battery cell based on a feature vector identified according to the first model.

[0043] According to an embodiment, the operation of identifying a first input voltage curve by supplying a current with a first waveform to a battery cell or identifying a second input voltage curve based on a current with a second waveform output from a battery cell may include the operation of identifying the first input voltage curve when the charging level of the battery cell falls within a pre-specified state range, and the operation of identifying the second input voltage curve when the charging level of the battery cell does not fall within the pre-specified state range.

[0044] According to an embodiment, the pre-specified state range may include values ​​smaller than the pre-specified state value.

[0045] According to an embodiment, the operation of diagnosing the state of a battery cell based on a feature vector identified according to a first model may include the operation of identifying the feature vector based on at least a component of a potential vector that is the result of processing in a bottleneck layer included in the first model.

[0046] According to an embodiment, the operation of identifying a feature vector based on at least a portion of the components of a potential vector as a result of processing in a bottleneck layer included in a first model may include diagnosing the state of a battery cell based on at least a portion of the feature vector, at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof; or diagnosing the state of a battery cell based on at least a portion of the feature vector, at least a portion of a second input voltage curve, at least a portion of a recovery voltage curve recovered based on a first model, at least a portion of the derivative of the second input voltage curve, or any combination thereof.

[0047] According to an embodiment, the operation of diagnosing the state of a battery cell based on a feature vector identified according to a first model may include the operation of inputting at least a portion of the feature vector into a second model that outputs the state of the battery cell.

[0048] According to an embodiment, the first model may include a Long Short-Term Memory Autoencoder (LSTM AE) model.

[0049] According to yet another embodiment of this document, a computer-readable recording medium records a program for performing a battery diagnostic method on a computer.

[0050] Beneficial effects

[0051] This technique can reduce the noise generation rate by using dimensionality-reduced feature vectors identified in the first model that is encoded and then recovered.

[0052] Furthermore, this technique can reduce the overfitting rate in the second model, which serves as a regression model, by reducing the noise occurrence rate.

[0053] Furthermore, this technology can improve the diagnostic accuracy of battery cell states based on feature vectors extracted by a first model that has already undergone a pattern learning process.

[0054] Furthermore, this technique can reduce the decrease in training stability caused by high-dimensional data processing.

[0055] Furthermore, this technique can improve training efficiency even when the amount of input data is less than a specified amount.

[0056] In addition, it can provide various effects that are directly or indirectly identified through this document. Attached Figure Description

[0057] Figure 1 This is a block diagram illustrating a battery pack in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0058] Figure 2 This is a block diagram illustrating the configuration of a battery diagnostic device and a battery diagnostic method according to embodiments of this document.

[0059] Figure 3 An example of executing code and then restoring a first model in a battery diagnostic device and battery diagnostic method according to embodiments of this document is illustrated.

[0060] Figure 4 The illustration shows an example of the operation of processing a voltage curve by performing encoding and then restoring a first model in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0061] Figure 5 The illustration shows an example of the operation of diagnosing the lifespan information of a battery cell based on the charge level of the battery cell in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0062] Figure 6 The illustration shows an example of an operation in a battery diagnostic device and battery diagnostic method according to embodiments of this document, wherein the data type input to a second model, which serves as a regression model, is changed according to the components of a latent vector.

[0063] Figure 7 The illustration shows an example of an operation in a battery diagnostic device and battery diagnostic method according to embodiments of this document, wherein the values ​​of data input into a second model, which is a regression model, are changed according to the components of a latent vector.

[0064] Figure 8 The illustration shows an example of how the method for deriving the state of a battery cell from data output as a second model, which is a regression model, changes according to the components of a potential vector in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0065] Figure 9 The illustration shows an example of the operation of diagnosing a single battery cell in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0066] Figure 10The illustration shows an example of the operation of a battery diagnostic device and battery diagnostic method according to embodiments of this document, which diagnoses the state of a battery cell based on a voltage curve of the battery cell obtained in a manner determined by the degree of charging of the battery cell.

[0067] Figure 11 This is a block diagram illustrating the hardware configuration of the computing system executing the battery diagnostic method in the battery diagnostic device and battery diagnostic method according to embodiments of this document. Detailed Implementation

[0068] In the following description, some embodiments described in this document are illustrated with reference to the accompanying drawings. However, this is not intended to limit the technology to the specific embodiments, but should be understood to include various modifications, equivalents, and / or substitutions of the embodiments incorporating the technology.

[0069] When adding reference numerals to components in each of the accompanying drawings, it should be noted that even if the same component is shown in different drawings, the same component is given the same reference numerals as much as possible. Furthermore, in describing the various embodiments disclosed in this document, detailed descriptions of relevant known configurations or functions are omitted if it is determined that such detailed descriptions would impede understanding of the embodiments of this disclosure. The singular form of a noun corresponding to an item may include one or more items unless the relevant context clearly indicates otherwise.

[0070] In describing the components of the embodiments described in this document, terms such as first, second, A, B, (a), (b), etc., may be used. These terms are intended only to distinguish components from other components, and the nature, order, or sequence of components is not limited by these terms. Furthermore, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments disclosed in this document pertain. Terms defined in common dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant art, and should not be interpreted in an ideal or overly formal sense unless expressly defined in this application.

[0071] Furthermore, in this disclosure, the expressions "greater than" or "less than" may be used to determine whether a specific condition is met or reached, but these are merely illustrative descriptions and do not exclude descriptions of "greater than or equal to" or "less than or equal to". A condition described as "greater than or equal to" may be replaced with "greater than", a condition described as "less than or equal to" may be replaced with "less than", and a condition described as "greater than or equal to and less than" may be replaced with "greater than and less than or equal to". Additionally, in the following text, "A" to "B" means at least one of the elements from A (inclusive) to B (inclusive).

[0072] In this document, each of the phrases “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” can include any one of the items listed together in the corresponding phrase or all possible combinations thereof.

[0073] In this document, when a component (e.g., a first component) is referred to, or not referred to, by the terms “functionally” or “communically” as “connected,” “coupled,” or “joined” to another component (e.g., a second component), it means that the component can be connected to the other component directly (e.g., via a wired connection), wirelessly, or via a third component.

[0074] Methods according to the various embodiments disclosed in this document can be provided by including them in a computer program product. The computer program product can be traded as a product between a seller and a buyer. The computer program product can be distributed in the form of a machine-readable recording medium (e.g., an optical disc read-only memory (CD-ROM)), or distributed through an app store, directly between two user devices, or distributed online (e.g., downloaded or uploaded). In the case of online distribution, at least a portion of the computer program product can be temporarily stored or temporarily generated in the memory of a machine-readable recording medium, such as a manufacturer's server, an app store server, or a relay server.

[0075] According to various embodiments, each of the above-described components (e.g., modules or programs) may include one or more entities, and some of the multiple entities may be separated and placed in other components. According to various embodiments, one or more of the above-described components or operations 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 in the same or similar manner as the functions performed by corresponding components among the multiple components prior to integration. According to various embodiments, operations performed by modules, programs, or other components may be performed sequentially, in parallel, repeatedly, or heuristically, or one or more operations may be performed in a different order, omitted, or performed by adding one or more other operations.

[0076] In the following text, reference will be made to Figures 1 to 11 The embodiments described in this document are described in detail.

[0077] Figure 1 This is a block diagram illustrating a battery pack in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0078] refer to Figure 1The battery pack 1 may include battery cells 12, sensor units 14, switching units 16, and a battery management system (BMS) 20. In this case, the battery pack 1 may be equipped with multiple battery cells 12, sensor units 14, switching units 16, and battery management systems 20.

[0079] According to an embodiment, battery cell 12 can supply power to a target device (not shown). For this purpose, battery cell 12 can be electrically connected to the target device. Here, the target device can include electrical, electronic, or mechanical devices that operate by receiving power from battery pack 1. For example, the target device can be, but is not limited to, an electric vehicle (EV) or an energy storage system (ESS).

[0080] According to an embodiment, the battery cell 12 may include at least one rechargeable and dischargeable battery cell 10. Here, the battery cell 10 may be a basic unit of a battery cell that can be used to generate electrical energy through charging and discharging. For example, the battery cell 10 may be a lithium-ion (Li-ion) battery, a lithium-ion polymer battery, a nickel-cadmium (Ni-Cd) battery, a nickel metal hydride (Ni-MH) battery, etc., but is not limited thereto.

[0081] According to the embodiments, multiple battery cells 12 can be connected in series or in parallel. For example, battery cell 12 can be a battery module, a battery bank, or a group of battery cells (battery-to-group structure).

[0082] According to an embodiment, sensor unit 14 can acquire information related to battery cell 12. According to an embodiment, sensor unit 14 can acquire values ​​(or information) related to the state of each of battery cell 12 or battery cell 10. In an embodiment, the state-related values ​​may include one or more values ​​of the battery cell's voltage, current, resistance, state of charge (SOC), state of health (SOH), or temperature, or combinations thereof.

[0083] According to an embodiment, sensor unit 14 can provide battery management system 20 with information about each individual battery cell in the plurality of battery cells 12.

[0084] According to an embodiment, the switching unit 16 may include a device for controlling the current for charging or discharging the battery cell 12. For example, depending on the specifications of the battery pack 1, the switching unit 16 may include at least one relay and / or magnetic contactor, etc.

[0085] According to an embodiment, the battery management system (BMS) 20 can control or manage the battery pack 1 to prevent overcharging and over-discharging by monitoring the voltage, current, temperature, etc. of the battery pack 1. For example, the battery management system 20 is an interface that receives values ​​obtained by measuring the various parameters mentioned above, and may include multiple terminals, circuitry connected to these terminals to process the received values, etc. Furthermore, the battery management system 20 can control the sensor unit 14 and / or the switching unit 16. For example, the battery management system 20 can be connected to multiple battery cells 12 to monitor the state of each individual battery cell in the multiple battery cells 12 and control the on / off state of relays or contactors, etc.

[0086] According to an embodiment, the operation of the battery management system 20 can be performed by the battery management system (BMS) in the vehicle, and can also be performed in various devices such as servers, cloud, chargers, or chargers / dischargers.

[0087] The upper-level controller 2 can transmit control signals for the multiple battery cells 12 to the battery management system 20. Therefore, the operation of the battery management system 20 can be controlled based on the signals applied from the upper-level controller 2.

[0088] According to an embodiment, the battery management system 20 may include Figure 2 The battery diagnostic device 201. According to another embodiment, the battery management system 20 can be integrated with... Figure 2 Battery diagnostic equipment 201 different systems. That is to say, Figure 2 The battery diagnostic device 201 can be included in the battery pack 1 or configured as another device outside the battery pack 1. In the following description, for ease of description, it will be based on the assumption that the battery diagnostic device 201 is composed of another device outside the battery pack 1. Furthermore, the operation of the battery diagnostic device 201 described below can be performed by the BMS in the vehicle, and can also be performed by various devices such as servers, cloud, chargers, and dischargers.

[0089] Figure 2 This is a block diagram illustrating the configuration of a battery diagnostic device and a battery diagnostic method according to embodiments of this document.

[0090] refer to Figure 2 The battery diagnostic device 201 may include a memory 203 and at least one processor 205. The memory 203 may store at least one instruction. The at least one processor 205 may execute at least one instruction.

[0091] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can identify the voltage curve of a battery cell by supplying a current with a first waveform to the battery cell, or by identifying the voltage curve of the battery cell based on a second waveform output from the battery cell, and diagnose the state of the battery cell (e.g., battery cell life information) based on the identified voltage curve.

[0092] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can diagnose the state of a single battery cell based on an input voltage curve and values ​​obtained based on a first model to improve diagnostic accuracy. In this case, the first model can be encoded and then the input voltage curve can be recovered.

[0093] According to an embodiment, the first model, which is encoded and then recovered, can be used to improve the diagnostic accuracy of the state of a battery cell (e.g., battery cell lifetime information). The first model may be referred to as a Long Short-Term Memory Automatic Encoder (LSTM-AE), but the embodiments in this document are not limited to this.

[0094] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can input the identified voltage curve as an input voltage curve into a first model.

[0095] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can input at least one of the following into a second model as a regression model: at least one portion of a feature vector identified based on a first model, at least one portion of an input voltage curve, at least one portion of an output voltage curve recovered based on the first model, at least one portion of the derivative of the input voltage curve, or any combination thereof.

[0096] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can diagnose the state of a battery cell (e.g., battery cell lifespan information) based on values ​​output from a second model, which is a regression model. (See below for reference.) Figure 3 and Figure 4 Describe the specific details.

[0097] The lifespan information of a battery cell may include the state of charge (SOH) of the battery cell, and the SOH represents the ratio of the current maximum charge capacity to the initial design capacity of the battery cell, and can be expressed as a value between 0 and 1, or a value between 0% and 100%. Methods such as cycle count-based assessments, internal resistance change analysis, and / or capacity degradation rate calculations can be used for SOH measurement, but the embodiments in this document are not limited to these.

[0098] Figure 3 An example of a first model in the battery diagnostic device and battery diagnostic method according to embodiments of this document is illustrated.

[0099] refer to Figure 3 The first model may include an autoencoder using LSTM, but the embodiments described herein are not limited to this. The first model may receive an input voltage curve as time-series data, extract feature vectors from the data, and reconstruct the input voltage curve as input data based on the feature vectors. Here, LSTM is one type of artificial neural network and can be defined as a structure that has been improved to allow recurrent neural networks (RNNs) to reflect long-term characteristics.

[0100] According to an embodiment, the first model may include an LSTM encoder, a bottleneck layer, and an LSTM decoder. However, the embodiments in this document are not limited thereto, and the first model without using LSTM may include an encoder, a bottleneck layer, and a decoder.

[0101] According to an embodiment, the first model can obtain feature vectors through a bottleneck layer, where the input data (e.g., an input voltage curve) is encoded by an LSTM encoder. Compared to the input data, the feature vectors can be lower-dimensional data.

[0102] According to an embodiment, the first model can obtain recovered data (e.g., recovered voltage curves) by decoding and recovering the feature vector in which the input data is compressed.

[0103] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can diagnose the state of a single battery cell by inputting at least one of input data, recovered data, feature vectors, or combinations thereof into a second model as a regression model, and diagnosing the state of a single battery cell based on the output value of the second model.

[0104] The second model can include not only linear regression models and kernel regression models, but also various other types of regression models.

[0105] According to an embodiment, the battery diagnostic device 201 may consist of a data collection unit, a feature extraction unit, and a lifespan prediction unit. The feature extraction unit may be configured as a first model. The lifespan prediction unit may be configured as a second model. The second model may identify the battery's state (e.g., battery lifespan information) based on parameters identified by the feature extraction unit.

[0106] Figure 4 The illustration shows an example of the operation of processing voltage curves using a first model in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0107] refer to Figure 4The input voltage curve set 401 can represent at least one voltage curve measured from each of at least one battery cell included in the battery cell. The input voltage curve 403 included in the input voltage curve set 401 can include the voltage curve of a first battery cell among the first to nth battery cells included in the battery cell. The input voltage curve 403 can include a vector representing the voltage of the first battery cell measured over time. For example, the input voltage curve 403 can include a vector representing the voltage measured at specified time intervals (e.g., approximately 0.1 seconds) within a specified time period (e.g., approximately 60 seconds).

[0108] The feature vector set 405 can represent at least one latent vector, each of which is the result of processing each input voltage curve in the bottleneck layer, which is included in the input voltage curve set 401. Feature vector 407 can be identified based on at least a portion of the components of the latent vectors of the first model for input voltage curve 403. The number of components of the feature vector (e.g., feature vector 407) can be less than the number of components of the input voltage curve (e.g., input voltage curve 403).

[0109] The set of recovery voltage curves 409 can represent at least one recovery voltage curve identified by decoding each feature vector in at least one feature vector included in the set of feature vectors 405. The set of recovery voltage curves 411 can represent the recovery voltage curve identified by decoding the latent vector 407.

[0110] According to an embodiment, the voltage curve of the first battery cell may include a first input voltage curve identified by supplying a current with a first waveform to the battery cell within a specified time period, or a second input voltage curve identified based on a current with a second waveform output from the battery cell.

[0111] In this case, it is possible to determine which voltage curve, the first or the second, to identify, is based on the charging level of the individual battery cells.

[0112] For example, when the charge level of a battery cell (e.g., the state of charge (SOC) of the battery cell) falls within a pre-specified range (e.g., less than about 50% SOC), at least one processor 205 of the battery diagnostic device 201 can identify a first input voltage curve by supplying a current with a first waveform to the battery cell over a specified time period.

[0113] This is because, when the charge level of a battery cell is below a predetermined state value, the stability of the voltage curve obtained by charging the battery cell is higher than the stability of the voltage curve obtained by discharging the battery cell. Furthermore, this is because, when the charge level of a battery cell is below a predetermined state value, the current that can be supplied to the battery cell during charging is greater than the current that the battery cell can output during discharging. Therefore, when the charge level of a battery cell is below a predetermined state value, the diagnostic accuracy of a method for diagnosing the battery state (e.g., battery life information, such as SOH) based on a first input voltage curve can be higher than the diagnostic accuracy of a method for diagnosing the battery state based on a second input voltage curve.

[0114] For example, when the charge level of a battery cell (e.g., the state of charge of the battery cell) does not fall within a pre-specified state range (e.g., the state of charge of the battery cell is less than about 50%), at least one processor 205 of the battery diagnostic device 201 can identify a second input voltage curve based on the current having a second waveform output from the battery cell. The first waveform and the second waveform can be the same or different.

[0115] This is because, when the charge level of a battery cell exceeds a predetermined state value, the stability of the voltage curve obtained by discharging the battery cell is higher than that obtained by charging the battery cell. Furthermore, this is because, when the charge level of a battery cell exceeds a predetermined state value, the current that the battery cell can output during discharge is greater than the current that can be supplied to the battery cell during charging. Therefore, when the charge level of a battery cell exceeds a predetermined state value, the diagnostic accuracy of a method for diagnosing the battery state (e.g., battery life information, such as SOH) based on a second input voltage curve can be higher than the diagnostic accuracy of a method for diagnosing the battery state based on a first input voltage curve.

[0116] Therefore, at least one processor 205 of the battery diagnostic device 201 can diagnose the state of a single battery cell, regardless of its charge level. Furthermore, since at least one processor 205 of the battery diagnostic device 201 identifies the input voltage curve (e.g., a first input voltage curve or a second input voltage curve) within a specified time period, the input voltage curve can be identified in a shorter time compared to the time required for identification in conventional battery diagnostic methods, and battery diagnostics can be performed in a shorter time compared to the time required in conventional battery diagnostic methods.

[0117] According to an embodiment, at least one processor 205 of the battery diagnostic device 201 can identify the state of a battery cell (e.g., the state of health of the battery cell) based on at least one of the following: an input voltage curve 403 (e.g., a first input voltage curve or a second input voltage curve), at least a portion of a feature vector 407, at least a portion of a recovery voltage curve 411, at least a portion of the derivative of the input voltage curve 403, or any combination thereof.

[0118] As a second model of the regression model, at least a portion of the input voltage curve 403, at least a portion of the eigenvector 407, at least a portion of the recovery voltage curve 411, at least a portion of the derivative of the input voltage curve 403, or any combination thereof, can be used to output parameters representing the state of the battery cell (e.g., the SOH of the battery cell) as continuous values ​​based on a regression equation such as Equation 1.

[0119] [Equation 1]

[0120] SOH can represent the lifespan information of a single battery cell. , and The regression weights can be determined by training a second model. , and It may include at least a portion of the input voltage curve 403, at least a portion of the eigenvector 407, at least a portion of the recovery voltage curve 411, at least a portion of the derivative of the input voltage curve 403, or at least one combination thereof. Parameters included in the regression equation may be pre-specified. Furthermore, Equation 1 is in the form of multiple linear equations, but the regression equations included in the embodiments of this document are not limited to this.

[0121] Figure 5 The illustration shows an example of the operation of diagnosing the lifespan information of a battery cell based on the charge level of the battery cell in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0122] In the following text, it is assumed that... Figure 5 At least one processor 205 of the battery diagnostic device 201 performs Figure 5 The process. Additionally, in Figure 5 In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0123] refer to Figure 5In the first operation (501), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify whether the charge level of a battery cell falls within a pre-specified state range (e.g., less than about 50% SOC). When the charge level of a battery cell falls within the pre-specified state range, at least one processor 205 of the battery diagnostic device 201 can perform a second operation (503). When the charge level of a battery cell does not fall within the pre-specified state range, at least one processor 205 of the battery diagnostic device 201 can perform a third operation (517).

[0124] In the second operation (503), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can supply a current having a first waveform to the battery cell.

[0125] In the fourth operation (505), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify the first input voltage curve. For example, a data collection unit included in the battery diagnostic device 201 can perform the fourth operation (505).

[0126] In the fifth operation (507), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can normalize the first input voltage curve.

[0127] For example, at least one processor 205 of the battery diagnostic device 201 can normalize the first input voltage curve by dividing the value obtained by subtracting the minimum value of the first input voltage curve from the first input voltage curve by the value obtained by subtracting the minimum value of the first input voltage curve from the maximum value of the first input voltage curve.

[0128] For example, the feature extraction unit included in the battery diagnostic device 201 can perform the fifth operation (507).

[0129] In the sixth operation (509), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can input a normalized first input voltage curve into a first model. The first model may include a model that performs encoding and then recovers.

[0130] For example, the feature extraction unit included in the battery diagnostic device 201 can perform the sixth operation (509).

[0131] In the seventh operation (511), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify feature vectors and recover voltage curves based on the first model.

[0132] For example, the feature extraction unit included in the battery diagnostic device 201 can perform the seventh operation (511).

[0133] In the eighth operation (513), at least one processor 205 of the battery diagnostic device 201 according to the embodiment may input at least one of the following into a second model: at least a portion of a feature vector, at least a portion of a first input voltage curve, at least a portion of a recovery voltage curve, at least a portion of the derivative of the first input voltage curve, or any combination thereof. The second model may include a regression model.

[0134] For example, the types of parameters input into the second model can be pre-specified or determined according to pre-specified criteria. (Reference) Figures 6 to 7 This describes an example of a standard used to determine the type of a parameter.

[0135] For example, the curves input into the second model can be determined based on pre-specified criteria. For example, at least one processor 205 of the battery diagnostic device 201 can input at least one component of the feature vector in a pre-specified order, at least one component of the first input voltage curve in a pre-specified order, at least one component of the recovery voltage curve in a pre-specified order, at least one value of at least one derivative of the first input voltage curve in a pre-specified order, or any combination thereof, into the second model.

[0136] For example, the life prediction unit included in the battery diagnostic device 201 can perform the eighth operation (513).

[0137] In the ninth operation (515), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the SOH of the battery cell based on the second model.

[0138] For example, the life prediction unit included in the battery diagnostic device 201 can perform the ninth operation (515).

[0139] In the third operation (517), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can output a current with a second waveform from a battery cell.

[0140] In the tenth operation (519), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify the second input voltage curve.

[0141] For example, the data collection unit included in the battery diagnostic device 201 can perform a tenth operation (519).

[0142] In the eleventh operation (521), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can normalize the second input voltage curve.

[0143] For example, in the fifth operation (507), at least one processor 205 of the battery diagnostic device 201 can normalize the second input voltage curve by dividing the value obtained by subtracting the minimum value of the second input voltage curve from the second input voltage curve by the value obtained by subtracting the minimum value of the second input voltage curve from the maximum value of the second input voltage curve.

[0144] For example, the feature extraction unit included in the battery diagnostic device 201 can perform an eleventh operation (521).

[0145] In the twelfth operation (523), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can input the second input voltage curve into the first model.

[0146] For example, the feature extraction unit included in the battery diagnostic device 201 can perform the twelfth operation (523).

[0147] In the thirteenth operation (525), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify feature vectors and recovery voltage curves based on the first model.

[0148] For example, the feature extraction unit included in the battery diagnostic device 201 can perform a thirteenth operation (525).

[0149] In the fourteenth operation (527), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can input at least one of the following into the second model: at least one portion of the feature vector, at least one portion of the second input voltage curve, at least one portion of the recovery voltage curve, at least one portion of the derivative of the second input voltage curve, or any combination thereof.

[0150] For example, as in the eighth operation (513), refer to Figure 6 and Figure 7 Examples of standards used to determine the type of parameters.

[0151] For example, in the eighth operation (513), the curve input into the second model can be determined according to a pre-specified standard.

[0152] For example, the life prediction unit included in the battery diagnostic device 201 can perform the fourteenth operation (527).

[0153] In the fifteenth operation (529), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the SOH of the battery cell based on the second model.

[0154] For example, the life prediction unit included in the battery diagnostic device 201 can perform the fifteenth operation (529).

[0155] Figure 6 The illustration shows an example of an operation in a battery diagnostic device and battery diagnostic method according to embodiments of this document, wherein the data type input to a second model, which serves as a regression model, is changed according to the components of a latent vector.

[0156] In the following text, it is assumed that... Figure 6 At least one processor 205 of the battery diagnostic device 201 performs Figure 6 The process. Additionally, in Figure 6 In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0157] The first operation (601), the second operation (603), and the third operation (605) can replace Figure 5 The eighth operation (513) and the ninth operation (515) can be performed, or the fourteenth operation (527) and the fifteenth operation (529) can be performed instead.

[0158] refer to Figure 6 In the first operation (601), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify whether all components of the feature vector fall within a pre-specified range (e.g., a range smaller than a pre-specified feature value). When all components of the feature vector fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 can perform a second operation (603). When any component of the feature vector does not fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 can perform a third operation (605).

[0159] In the second operation (603), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the first SOH of the battery cell by inputting at least a portion of the input voltage curve (e.g., the first input voltage curve or the second input voltage curve), at least a portion of the recovery voltage curve (e.g., the recovery voltage curve of the first input voltage curve or the recovery voltage curve of the second input voltage curve), at least a portion of the derivative of the input voltage curve or any combination thereof, and at least a portion of the feature vector into the second model when all components of the feature vector fall within a pre-specified range.

[0160] In the third operation (605), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the second SOH of the battery cell by inputting at least a portion of the input voltage curve (e.g., the first input voltage curve or the second input voltage curve), at least a portion of the recovery voltage curve (e.g., the recovery voltage curve of the first input voltage curve or the recovery voltage curve of the second input voltage curve), at least a portion of the derivative of the input voltage curve, or any combination thereof, into the second model.

[0161] In other words, noise may occur in the feature vector when at least one of the components of the feature vector does not fall within a pre-specified range, and therefore, at least one processor 205 of the battery diagnostic device 201 can identify the life information of the battery cell (e.g., second SOH) by inputting at least one parameter other than the feature vector into the second model to reduce the impact of noise.

[0162] According to the embodiment, during execution Figure 6 In the battery diagnostic device 201, the type of parameters input into the second model can change as the components of the feature vector gradually increase or decrease.

[0163] Figure 7 The illustration shows an example of an operation in a battery diagnostic device and battery diagnostic method according to embodiments of this document, wherein the values ​​of data input into a second model, which is a regression model, are changed according to the components of a latent vector.

[0164] In the following text, it is assumed that... Figure 7 At least one processor 205 of the battery diagnostic device 201 performs Figure 7 The process. Additionally, in Figure 7 In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0165] The first operation (701), the second operation (703), and the third operation (705) can replace Figure 5 The eighth operation (513) and the ninth operation (515) can be performed, or the fourteenth operation (527) and the fifteenth operation (529) can be performed instead.

[0166] refer to Figure 7In the first operation (701), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify whether all components of the feature vector fall within a pre-specified range (e.g., a range smaller than a pre-specified feature value). When all components of the feature vector fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 can perform a second operation (703). When any component of the feature vector does not fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 according to the embodiment can perform a third operation (705).

[0167] In the second operation (703), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the first SOH of the battery cell by inputting at least a portion of the input voltage curve, at least a portion of the recovery voltage curve, at least a portion of the derivative of the input voltage curve or any combination thereof, and at least a portion of the feature vector into the second model.

[0168] In the third operation (705), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can replace any component of the feature vector that does not fall within the pre-specified range with a boundary value of a pre-specified range.

[0169] In other words, since noise may appear in the feature vector, at least one processor 205 of the battery diagnostic device 201 can replace any component of the feature vector that is greater than the upper limit of a pre-specified range with an upper limit value, and replace any component of the feature vector that is less than the lower limit of a pre-specified range with a lower limit value, in order to reduce the impact of noise. Thereafter, at least one processor 205 of the battery diagnostic device 201 can perform a second operation (703) based on the feature vector with some components replaced.

[0170] According to the embodiment, during execution Figure 7 In the battery diagnostic device 201, as the components of the feature vector gradually increase or decrease, the parameters indicating the state of the battery cell (e.g., the state of health (SOH) of the battery cell) can converge to a specific value.

[0171] Figure 8 The illustration shows an example of how the method for deriving the state of a battery cell from data output as a second model, which is a regression model, changes according to the components of a potential vector in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0172] In the following text, it is assumed that... Figure 8 At least one processor 205 of the battery diagnostic device 201 performs Figure 8 The process. Additionally, in Figure 8In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0173] The first operation (801), the second operation (803), and the third operation (805) can replace Figure 5 The eighth operation (513) and the ninth operation (515) can be performed, or the fourteenth operation (527) and the fifteenth operation (529) can be performed instead.

[0174] In the first operation (801), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify whether all components of the feature vector fall within a pre-specified range (e.g., a range smaller than a pre-specified feature value). When all components of the feature vector fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 according to the embodiment can perform a second operation (803). When at least one component of the feature vector does not fall within the pre-specified range, at least one processor 205 of the battery diagnostic device 201 according to the embodiment can perform a third operation (805).

[0175] In the second operation (803), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the SOH of the battery cell based on the first SOH and the second SOH.

[0176] The first SOH can be represented as the SOH output from the second model by inputting at least a portion of the input voltage curve, at least a portion of the recovered voltage curve, at least a portion of the derivative of the input voltage curve, or any combination thereof, and at least one of the eigenvectors into the second model.

[0177] The second SOH can be represented as the SOH output from the second model by inputting at least a portion of the input voltage curve, at least a portion of the recovery voltage curve, at least a portion of the derivative of the input voltage curve, or any combination thereof into the second model.

[0178] For example, at least one processor 205 of the battery diagnostic device 201 can estimate the SOH of a single battery cell based on the average of the first SOH and the second SOH.

[0179] In the third operation (805), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can estimate the SOH of the battery cell based on the second SOH. In other words, since noise may appear in the feature vector, at least one processor 205 of the battery diagnostic device 201 can estimate the SOH of the battery cell using the second SOH to reduce the impact of noise.

[0180] According to the embodiment, during execution Figure 6 In the battery diagnostic device 201, as the components of the feature vector gradually increase or decrease, the type of parameters input into the second model and the method of deriving the state of the battery cell can be changed.

[0181] Figure 9 The illustration shows an example of the operation of diagnosing the state of a battery cell in a battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0182] In the following text, it is assumed that... Figure 9 At least one processor 205 of the battery diagnostic device 201 performs Figure 9 The process. Additionally, in Figure 9 In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0183] In the first operation (901), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify a first input voltage curve measured by supplying a current having a first waveform to a battery cell.

[0184] In the second operation (903), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can input a first input voltage curve into a first model. The first model may include a model that performs encoding and then recovers.

[0185] In the third operation (905), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can diagnose the state of a battery cell based on a feature vector identified according to the first model.

[0186] Figure 10 The illustration shows an example of the operation of a battery diagnostic device and battery diagnostic method according to embodiments of this document, which diagnoses the state of a battery cell based on a voltage curve of the battery cell obtained in a manner determined by the degree of charging of the battery cell.

[0187] In the following text, it is assumed that... Figure 10 At least one processor 205 of the battery diagnostic device 201 performs Figure 10 The process. Additionally, in Figure 10 In the description, the operations described as being performed by the battery diagnostic device 201 can be understood as being controlled by at least one processor 205 of the battery diagnostic device 201.

[0188] In the first operation (1001), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can identify a first input voltage curve by supplying a current having a first waveform to a battery cell, or identify a second input voltage curve based on a current having a second waveform output from a battery cell.

[0189] In the second operation (1003), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can input a first input voltage curve or a second input voltage curve into the first model.

[0190] In the third operation (1005), at least one processor 205 of the battery diagnostic device 201 according to the embodiment can diagnose the state of a battery cell based on a feature vector identified according to a first model.

[0191] Figure 11 This is a block diagram illustrating the hardware configuration of the computing system executing the battery diagnostic method in the battery diagnostic device and battery diagnostic method according to embodiments of this document.

[0192] refer to Figure 11 The computing system 1100 according to the embodiments disclosed in this document may include an MCU 1110, a memory 1120, an input / output interface 1130, and a communication interface 1140.

[0193] MCU 1110 can be one or more processors that execute various programs stored in memory 1220 (e.g., battery cell data collection programs, graphics generation programs, data analysis programs, data decomposition algorithms, normalization programs, and battery cell diagnostic programs, etc.). These programs process various information, including battery cell characteristic data and latent variables, and execute the aforementioned... Figures 2 to 10 The battery diagnostic device 101 shown has the following functions.

[0194] The memory 1120 can store various programs such as battery cell data collection programs, graphics generation programs, data analysis programs, data decomposition algorithms, normalization programs, and battery cell diagnostic programs.

[0195] Multiple such memories 1120 can be provided as needed. Memory 1120 can be volatile or non-volatile memory. Memory 1120 used as volatile memory can be RAM, DRAM, SRAM, etc. Memory 1120 used as non-volatile memory can be ROM, PROM, EAROM, EPROM, EEPROM, flash memory, etc. The examples of memories 1120 listed above are merely examples, but are not limited to these examples.

[0196] The Input / Output I / F 1130 provides an interface that allows data to be sent and received by connecting input devices (not shown) such as a keyboard, mouse, or touch panel and output devices (not shown) such as a display to the MCU 1110.

[0197] The communication I / F 1140 is configured to send and receive various data with a server and can be various devices that support wired or wireless communication. For example, the battery diagnostic device 201 can send and receive various information, including the shape model of a battery cell, to and from a separately provided external server via the communication I / F 1140.

[0198] In this way, a computer program according to the embodiments disclosed in this document can be implemented to execute, for example, by being recorded in memory 1120 and processed by MCU 1110. Figure 2 The modules for each function are shown.

[0199] In the foregoing, although all components constituting the embodiments disclosed in this document have been described as being combined or operating in combination, the embodiments disclosed in this document are not necessarily limited to such embodiments. That is, within the scope of the purposes of the embodiments disclosed in this document, all components may be selectively combined and operated in one or more combinations.

[0200] Furthermore, the terms "comprising," "configuration," or "having," unless otherwise expressly stated, mean that they may include the corresponding components and should therefore be interpreted as capable of further including rather than excluding other components. Unless otherwise defined, all terms including technical or scientific terms have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments disclosed in this document pertain. Commonly used terms, such as those defined in dictionaries, should be interpreted as consistent with the meaning in the context of the relevant art and should not be interpreted in an idealized or overly formal sense unless expressly defined in this document.

[0201] The foregoing disclosure outlines features of several embodiments, enabling those skilled in the art to better understand various aspects of this disclosure. Those skilled in the art will understand that this disclosure can readily serve as a basis for designing or modifying different structures to perform the same purpose or achieve the same advantages as the embodiments described in this document. Furthermore, those skilled in the art will recognize that such equivalent configurations do not depart from the scope of this disclosure, and that various changes, substitutions, and modifications can be made in this specification without departing from the scope of this disclosure.

Claims

1. A battery diagnostic device, comprising: A memory configured to store at least one instruction; At least one processor, the at least one processor being configured to execute the at least one instruction. Wherein, the at least one processor is configured to: Identification includes a first input voltage curve, which is measured by supplying a current with a specified waveform to a battery cell; The first input voltage curve is input into the first model, which encodes and then recovers the first input voltage curve; and The state of the battery cell is diagnosed based on the feature vectors identified by the first model.

2. The battery diagnostic device according to claim 1, wherein, The at least one processor is configured to identify the feature vector based on a potential vector that is a processing result in the bottleneck layer included in the first model.

3. The battery diagnostic device according to claim 2, wherein, The at least one processor is configured to diagnose the state of the battery cell based on at least a portion of the feature vector, at least a portion of the first input voltage curve, at least a portion of the recovery voltage curve recovered based on the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof.

4. The battery diagnostic device according to claim 2, wherein, The at least one processor is configured to: Based on determining that all components of the feature vector fall within a pre-specified range, the state of the battery cell is diagnosed according to at least one of at least a portion of the first input voltage curve, at least a portion of the recovery voltage curve recovered based on the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, and at least a portion of the feature vector. or The state of the battery cell is diagnosed based on at least one component of the feature vector not falling within the pre-specified range, based on at least a portion of the first input voltage curve, at least a portion of the recovered voltage curve recovered by the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof.

5. The battery diagnostic device according to claim 1, wherein, The at least one processor is configured to input at least a portion of the feature vector into a second model, the second model outputting the state of the battery cell.

6. The battery diagnostic device according to claim 1, wherein, The at least one processor is configured to supply the current to the battery cell during a specified time period.

7. The battery diagnostic device according to claim 1, wherein, The state of the battery cell includes a state of health (SOH).

8. The battery diagnostic device according to claim 1, wherein, The first model includes a Long Short-Term Memory Autoencoder (LSTM AE) model.

9. A battery diagnostic method, comprising: The identification process includes the operation of a first input voltage curve, which is measured by supplying a current with a specified waveform to a battery cell. The operation of inputting the first input voltage curve into the first model, the first model encoding the first input voltage curve and then recovering it; as well as The operation is based on diagnosing the state of the battery cell according to the feature vector identified by the first model.

10. The battery diagnostic method according to claim 9, wherein, The operation of diagnosing the state of the battery cell based on the feature vector identified according to the first model includes the operation of identifying the feature vector based on a latent vector that is a processing result in the bottleneck layer included in the first model.

11. The battery diagnostic method according to claim 10, wherein, The operation of diagnosing the state of the battery cell based on the feature vector identified according to the first model includes diagnosing the state of the battery cell based on at least a portion of the feature vector, at least a portion of the first input voltage curve, at least a portion of the recovery voltage curve recovered based on the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof.

12. The battery diagnostic method according to claim 10, wherein, The operation of identifying the feature vector based on the latent vector, which is the processing result of the bottleneck layer included in the first model, includes: The operation of diagnosing the state of a battery cell based on at least one of the following: determining that all components of the feature vector fall within a pre-specified range; based on at least one portion of the first input voltage curve, at least one portion of the recovered voltage curve recovered by the first model, at least one portion of the derivative of the first input voltage curve, or any combination thereof, and at least one portion of the feature vector; or The operation of diagnosing the state of the battery cell based on at least a portion of the first input voltage curve, at least a portion of the recovered voltage curve recovered by the first model, at least a portion of the derivative of the first input voltage curve, or any combination thereof, based on determining that at least one of the components of the feature vector does not fall within the pre-specified range.

13. The battery diagnostic method according to claim 9, wherein, The operation of diagnosing the state of the battery cell based on the feature vector identified by the first model includes inputting at least a portion of the feature vector into a second model, the second model outputting the state of the battery cell.

14. The battery diagnostic method according to claim 9, wherein, The identification includes the operation of a first input voltage curve measured by supplying a current with a specified waveform to a battery cell, and the operation of supplying the current to the battery cell within a specified time period.

15. The battery diagnostic method according to claim 9, wherein, The state of the battery cell includes a state of health (SOH).

16. The battery diagnostic method according to claim 9, wherein, The first model includes a Long Short-Term Memory Autoencoder (LSTM AE) model.

17. A computer-readable recording medium having a program recorded thereon for performing the method according to any one of claims 9 to 16 on a computer.

Images