Apparatus and method for diagnosing a battery

Abstract

An apparatus for diagnosing a battery includes a communication interface for receiving battery data obtained by applying a current signal to a battery module, and at least one processor for performing error correction to correct an influence of a sensing line related to a structure of the battery module and generating pre-processed data from the battery data, extracting a feature value based on the pre-processed data, and diagnosing whether the battery module is abnormal based on a plurality of feature values extracted at different time regions.

Description

Technical Field

[0001] Cross-references to related applications

[0002] This application claims priority to Korean Patent Application No. 10-2024-0067360, filed on May 23, 2024, and Korean Patent Application No. 10-2025-0049565, filed on April 16, 2025, the disclosures of which are incorporated herein by reference.

[0003] Field of the Invention

[0004] This invention relates to an apparatus and method for diagnosing batteries. Background Technology

[0005] Overall, research and development of rechargeable batteries has been actively pursued recently. Rechargeable batteries refer to batteries that can be charged and discharged, and include traditional Ni / Cd batteries, Ni / MH batteries, and more recently, lithium-ion batteries. Among rechargeable batteries, lithium-ion batteries have the following advantages: they have a much higher energy density than traditional Ni / Cd and Ni / MH batteries. Furthermore, lithium-ion batteries can be manufactured to be small and lightweight, and are therefore used as power sources for mobile devices. In addition, the applications of lithium-ion batteries have expanded to include powering electric vehicles, and they are attracting attention as a next-generation energy storage medium.

[0006] If defects exist in such batteries, heat or fire may be generated during charging, discharging, and use, potentially posing a safety risk. Therefore, research is underway on how to diagnose batteries based on battery data. In particular, there is a need to inspect battery packs over short periods to minimize downtime for battery-using devices and ensure their performance and safety, but there is a lack of relevant technologies. Summary of the Invention

[0007] Technical issues

[0008] One aspect of the present invention provides an apparatus for diagnosing a battery, which is capable of accurately analyzing voltage behavior by applying a current signal to the battery pack in order to quickly assess the state of the battery.

[0009] The technical problems of the embodiments disclosed in this document are not limited to those mentioned above, and those skilled in the art can clearly understand other unmentioned technical problems based on the following description.

[0010] Technical solution

[0011] According to one aspect of the present invention, an apparatus for diagnosing a battery is provided, the apparatus comprising: a communication interface for receiving battery data acquired by applying a current signal to a battery module; and at least one processor for performing error correction to correct for the influence of sensing lines related to the structure of the battery module and generating preprocessed data based on the battery data, extracting feature values ​​based on the preprocessed data, and diagnosing the state of the battery module based on a plurality of feature values ​​extracted at different time regions.

[0012] At least one processor can apply a linear transformation to the preprocessed data to obtain linearly transformed data, and extract feature values ​​based on the linearly transformed data.

[0013] At least one processor can group battery data according to the shape of the sensing line of the battery module to generate multiple sensing line group data, and generate preprocessed data by using correction values ​​to correct the differences between the multiple sensing line group data.

[0014] At least one processor can derive a correction value at the point where the difference between multiple sensing line group data is greatest, and use the correction value to correct the difference between the multiple sensing line group data.

[0015] At least one processor can derive a correction value based on the average of the correction target data and the average of the remaining data included in multiple sensing line group data.

[0016] At least one processor can derive a correction value from a value obtained by excluding the average of the mean of the correction target data and the average of the remaining data from the mean of the correction target data, and can use the correction value to correct for discrepancies.

[0017] At least one processor can diagnose the state of a battery module by applying weights determined based on the structure of the battery module to multiple feature values.

[0018] At least one processor can calculate the proximity of multiple battery cells included in a battery module and diagnose the state of the battery module by applying different weights to multiple feature values ​​based on the proximity.

[0019] At least one processor can diagnose the state of a battery module by applying different weights to multiple feature values ​​based on the relative positions of power lines connected to multiple battery cells and battery modules.

[0020] At least one processor can generate a trend line obtained by removing outliers based on the correlation between multiple feature values ​​derived from differences in chemical reactions occurring over time, and diagnose the state of the battery module based on the trend line.

[0021] At least one processor can export the distance value between each battery cell included in the battery module and the trend line, and diagnose battery cells whose distance value is greater than or equal to a preset reference value as abnormal.

[0022] According to another aspect of the present invention, a method for managing a battery is provided, the method comprising: receiving battery data acquired by applying a current signal to a battery module; performing error correction to correct for the influence of sensing lines related to the structure of the battery module and generating preprocessed data based on the battery data; extracting feature values ​​based on the preprocessed data; and diagnosing the state of the battery module based on a plurality of feature values ​​extracted at different time regions.

[0023] Feature extraction can include applying a linear transformation to the preprocessed data to obtain linearly transformed data, and then extracting feature values ​​based on the linearly transformed data.

[0024] Generating preprocessed data may include grouping battery data according to the shape of the sensing lines of the battery module to generate multiple sensing line groups of data, and generating preprocessed data by using correction values ​​to correct the differences between the multiple sensing line groups of data.

[0025] Correcting the differences between multiple sensor line group data can include deriving a correction value at the point where the difference between the multiple sensor line group data is the largest, and using the correction value to correct the differences between the multiple sensor line group data.

[0026] The correction value can be derived based on the average of the correction target data and the average of the remaining data included in multiple sensing line group data.

[0027] Exporting correction values ​​can include exporting the value obtained by excluding the average of the mean of the correction target data and the average of the remaining data from the mean of the correction target data.

[0028] Diagnosing the state of a battery module can include applying weights determined based on the battery module's structure to multiple feature values.

[0029] Diagnosing the state of a battery module can include calculating the proximity of multiple battery cells included in the battery module and diagnosing the state of the battery module by applying different weights to multiple feature values ​​based on the proximity.

[0030] Diagnosing the state of a battery module can include applying different weights to multiple feature values ​​based on the relative positions of the power lines connected to multiple battery cells and the battery module.

[0031] Diagnosing whether a battery module is abnormal may include generating a trend line obtained by removing outliers based on the correlation between multiple feature values ​​derived from differences in chemical reactions that occur over time, and diagnosing the state of the battery module based on this trend line.

[0032] Diagnosing the status of a battery module can include deriving the distance value between each battery cell in the module and the trend line, and diagnosing battery cells whose distance value is greater than or equal to a preset reference value as abnormal.

[0033] Beneficial effects

[0034] The apparatus for diagnosing a battery according to the embodiment can improve the reliability of the battery system by quickly diagnosing the state of the battery, and can improve efficiency by reducing the time associated with battery diagnosis. Attached Figure Description

[0035] Figure 1 This is a block diagram illustrating a general battery system including a device for diagnosing a battery according to an embodiment; Figure 2 This is a block diagram illustrating the configuration of a device for diagnosing batteries according to an embodiment; Figure 3 The flowchart illustrates a process for determining the state of a battery by means of a battery diagnostic apparatus according to an embodiment. Figure 4a This diagram shows a voltage curve of the battery voltage over time according to an embodiment. Figure 4b A linear transformation graph before error correction is shown in the apparatus for diagnosing batteries according to an embodiment. Figure 5 The diagram shows characteristic values ​​used in a battery diagnostic apparatus according to an embodiment before error correction. Figure 6 Other characteristic values ​​prior to error correction are shown in the apparatus for diagnosing batteries according to an embodiment; Figure 7 The structure of a battery pack is shown, which is the diagnostic target of a device for diagnosing batteries according to an embodiment; Figure 8 A linear transformation graph after error correction is shown in the apparatus for diagnosing batteries according to an embodiment. Figure 9 The diagram illustrates the characteristic values ​​included in a discharge curve used in a device for diagnosing batteries according to an embodiment; Figure 10 The diagram illustrates the feature values ​​included in the linear transformation curve used in a device for diagnosing batteries according to an embodiment; Figure 11The feature values ​​included in the average curve graph used in the device for diagnosing batteries according to an embodiment are shown. Figure 12 A trend line generated by a battery diagnostic apparatus according to an embodiment is shown; Figure 13 A trend line generated by a battery diagnostic apparatus according to an embodiment is shown; Figure 14 A trend line generated by a battery diagnostic apparatus according to an embodiment is shown; Figure 15 A control flowchart illustrating a method for diagnosing a battery according to an embodiment is shown; and Figure 16 Shown in Figure 15 The following is a control flowchart of the method for diagnosing a battery according to an embodiment. Detailed Implementation

[0036] In the following, various embodiments disclosed in this document will be described in detail with reference to the accompanying drawings. Throughout this document, the same reference numerals are used for the same components in the drawings, and repeated descriptions of the same components are omitted.

[0037] The specific structural or functional descriptions of the various embodiments disclosed in this document are provided for the purpose of describing these embodiments only, and the various embodiments disclosed in this document can be implemented in various forms and should not be construed as being limited to the embodiments described in this document.

[0038] Expressions such as “first,” “second,” “firstly,” or “secondarily” used in various embodiments may modify various components regardless of their order and / or importance, and do not limit these components. For example, a first component may be named a second component without departing from the scope of the embodiments disclosed in this document, and similarly, a second component may be renamed a first component.

[0039] The terminology used in this document is for describing particular embodiments only and is not intended to limit the scope of other embodiments. Singular expressions may include plural expressions unless the context clearly indicates otherwise.

[0040] All terms used herein, including technical or scientific terms, may have the same meaning as commonly understood by one of ordinary skill in the art as to the embodiments disclosed in this document. Terms as defined in common dictionaries may be interpreted as having the same or similar meaning in the context of related art, and should not be interpreted in an idealized or overly formal sense unless clearly defined in this document. In some cases, even terms defined in this document should not be construed as excluding the embodiments disclosed in this document.

[0041] Figure 1 This is a block diagram illustrating the configuration of a general battery system including devices for diagnosing batteries according to various embodiments.

[0042] Specifically, Figure 1 A battery system 10 and a higher-level controller 20 included in a higher-level system are schematically illustrated according to an embodiment disclosed in this document.

[0043] like Figure 1 As shown, the battery system 10 may include multiple battery modules 12, sensor units 14, switching units 16, and a device 1 for diagnosing the battery. In this case, the battery system 10 may be equipped with multiple battery modules 12, multiple sensor units 14, multiple switching units 16, and multiple devices 1 for diagnosing the battery.

[0044] Each of the plurality of battery modules 12 may include at least one battery cell 13 capable of being charged and discharged. The battery cell 13 may include a positive electrode, a positive electrode material, a negative electrode, a negative electrode material, a separator, an electrolyte, a polymer, and a casing. In this configuration, the plurality of battery modules 12 may be connected in series or in parallel.

[0045] Additionally, the battery system 10 may include a battery pack, which can be configured by combining multiple battery modules 12, and may include a battery management system (BMS) to monitor and control the battery state. Furthermore, the battery system 10 may include a battery bank, which can be configured as an energy storage system by combining multiple battery packs.

[0046] The sensor unit 14 may include a current sensor, a voltage sensor, and a temperature sensor.

[0047] The current sensor can detect the current used during the process of determining the SOC of the battery cell 13.

[0048] The current sensor can measure the battery current flowing in the battery, i.e., the charging current and the discharging current, and send the measurement results to the device 1 for diagnosing the battery. According to an embodiment, the current sensor can measure the battery current at a predetermined cycle during a charging cycle in which the battery is charged with power from an external device 3 or during a discharging cycle in which the battery is discharged, and send the measurement results to the device 1 for diagnosing the battery.

[0049] A voltage sensor can be configured to be connected in parallel with a battery to detect the battery voltage (which is the voltage across the two ends of the battery) and generate a voltage signal representing the detected battery voltage.

[0050] A temperature sensor can be configured to measure battery temperature and generate a temperature signal representing the measured battery temperature. The temperature sensor can be placed inside a housing to measure a temperature close to the actual temperature of the battery. For example, the temperature sensor can be attached to the surface of at least one battery cell included in a cell group, and the surface temperature of the battery cell can be detected as the battery temperature.

[0051] The switching unit 16 can be connected in series to the positive (+) terminal or the negative (-) terminal of the battery module 12 to control the charging / discharging current flow of the battery module 12. For example, depending on the specifications of the battery system 10, at least one relay, magnetic contactor, etc., can be used as the switching unit 16.

[0052] The device 1 for diagnosing the battery can monitor the voltage, current, temperature, etc. of the battery system 10, and control and manage the monitored voltage, current, temperature to prevent overcharging and over-discharging, etc., and may include, for example, a battery management system (BMS).

[0053] The device 1 for diagnosing batteries serves as an interface for receiving various parameters and may include multiple terminals and circuitry connected to these terminals and processing the received values. Additionally, the device 1 for diagnosing batteries may control the switching of switching units 16—e.g., relays or contactors—on / off and may be connected to battery modules 12 to monitor the status of each battery module 12.

[0054] In addition, the device 1 for diagnosing the battery can receive current data, voltage data and temperature data from the sensor unit 14 to obtain battery data and diagnose the state of the battery.

[0055] The upper-level controller 20 can send control signals to the device 1 for diagnosing the battery for controlling the battery module 12. Therefore, the operation of the device 1 for diagnosing the battery can be controlled based on the control signals applied from the upper-level controller 20. Additionally, the battery module 12 can be a component included in an energy storage system (ESS). In this case, the upper-level controller 20 can be a battery bank controller (BBMS) that includes multiple battery systems 10 or an ESS controller that controls the entire ESS including multiple battery banks. However, the battery system 10 is not limited to these uses.

[0056] Figure 2 This is a block diagram illustrating the configuration of a device for diagnosing batteries according to an embodiment.

[0057] refer to Figure 2According to an embodiment, the device 1 for diagnosing a battery may include: a control unit 100, which includes at least one processor 110 and a memory 120; and a communication interface 200, which communicates with an external device 3 through the communication interface 200 to diagnose the battery and sends the diagnostic results to the external device.

[0058] According to an embodiment, the device 1 for diagnosing a battery may include a user terminal and / or a server device capable of communicating with an external device 3.

[0059] Specifically, when the device 1 for diagnosing the battery is a user terminal, the control unit 100 of the device 1 for diagnosing the battery can be configured as the CPU of the user terminal to perform on-device diagnostics on the battery from the user terminal. In this case, the user terminal may include, but is not limited to, personal computers, terminals, mobile phones, smartphones, handheld devices, wearable devices, etc.

[0060] Furthermore, when the device 1 used for diagnosing the battery is a server device, the server device can be implemented as various computing devices, such as workstations, clouds, data drives, data stations, etc. The server device can be implemented as one or more server devices that are physically or logically separated based on functions, detailed configurations of functions, or data, and can send and receive data and process the sent and received data through communication between server devices.

[0061] The device 1 for diagnosing a battery according to the embodiment can refer to any electronic device including a processor 110 and a memory 120, and can be installed in a vehicle and operated. The components of the device 1 for diagnosing a battery will be described in detail below.

[0062] Figure 2 The battery data acquisition device 2 shown may include a separate, physically configured data acquisition unit. Here, the data acquisition unit can refer to a device that uses sensors and measuring equipment to monitor the battery's status and performance in real time, store data, and acquire and analyze data.

[0063] Furthermore, the battery data acquisition device 2 can be configured to operate within a device for diagnosing the battery. That is, the battery data acquisition device 2 can operate as a hardware configuration within the device for diagnosing the battery or it can operate in software form to acquire battery data.

[0064] Here, the battery data acquisition device 2 may include a constant current power supply for applying a constant current signal and a data acquisition device (DAQ).

[0065] The communication interface 200 may include a wireless communication interface 210 and a wired communication interface 220 for communicating with the external device 3. The communication interface 200 may send programs or various data for calculating the characteristic values ​​of battery cells, classifying them by category, and estimating their lifespan to a separately provided external server, and receive programs or various data for calculating the characteristic values ​​of battery cells, classifying them by category, and estimating their lifespan from the separately provided external server.

[0066] The wireless communication interface 210 may include at least one of a short-range communication module and a long-range communication module.

[0067] The short-range communication module can communicate with an external device 3 adjacent to the device 1 used for battery diagnostics using a short-range communication method. Here, the short-range communication module can utilize one of the following communication methods: Bluetooth, Bluetooth Low Energy (BLE), Infrared Data Association (IrDA), Zigbee, Wi-Fi, Wi-Fi Direct, Ultra Wideband (UWB), or Near Field Communication (NFC).

[0068] The long-distance communication module may include communication modules that perform various types of long-distance communication and may include a mobile communication interface. This mobile communication interface can transmit wireless signals to and receive wireless signals from at least one of the base station, external terminal, and external device 3 on a mobile communication network. Additionally, the long-distance communication module can communicate with external device 3 or other electronic devices via a nearby access point (AP). The access point (AP) can connect the local area network (LAN) to which the device 1 for battery diagnostics is connected to the wide area network (WAN) to which the communication server is connected. Accordingly, the device 1 for battery diagnostics can connect to the communication server via the external device 3 and the wide area network (WAN) to communicate with each other.

[0069] The wired communication interface 220 can be connected to a wired communication network and communicate with the external device 3 through the wired communication network. For example, the wired communication interface 220 can be connected to the wired communication network via Ethernet (IEEE 802.3 technical standard) or via CAN communication, and can send data to and receive data from the external device 3 through the wired communication network.

[0070] The device 1 for diagnosing batteries according to an embodiment may include an input / output interface (not shown). This interface allows data to be sent and received by interconnecting input devices (not shown), such as a keyboard, mouse, and touch panel, and output devices (not shown), such as a display, with the processor 110.

[0071] The memory 120 can store various information required for operating the device 1 for diagnosing the battery. Specifically, the memory 120 can store the operating system and programs required for operating the device 1 for diagnosing the battery, or it can store the data required for operating the device 1 for diagnosing the battery. In addition, the memory 120 can also store machine learning-related learning models required for operating the device 1 for diagnosing the battery.

[0072] Specifically, the memory 120 can store various programs related to calculating the characteristic values ​​of individual battery cells, classifying them, and estimating their lifespan. Additionally, the memory 120 can store various data, such as the voltage, current, temperature, and characteristic value data of each individual battery cell.

[0073] In addition, the memory 120 can store the diagnostic results of the battery cells diagnosed by the processor 110, as well as the specific information of the battery cells.

[0074] The memory 120 may include volatile memory 120, such as static random access memory (S-RAM) and dynamic random access memory (D-RAM) for temporary data storage. In addition, the memory 120 may also include non-volatile memory 120, such as read-only memory (ROM), erasable programmable read-only memory (EPROM), and electrically erasable programmable read-only memory (EEPROM) for storing data for long periods of time.

[0075] Processor 110 outputs control signals to control the device 1 used for diagnosing the battery as a whole. Processor 110 may include one or more central processing units (CPUs) and graphics processing units (GPUs). In this case, processor 110 may be implemented as an array of multiple logic gates, or it may be implemented as a combination of a general-purpose microprocessor 110 and a memory 120, which stores programs that can be executed on the microprocessor 110.

[0076] Specifically, the processor 110 can perform error correction to correct the influence of the sensing line on the battery data to generate preprocessed data, and here, error correction to correct the influence of the sensing line can mean the process of correcting errors that occur based on the physical structure of the battery pack.

[0077] Specifically, sensing lines are wirings designed during the battery pack design process, based on the structure of individual battery cells, battery modules, and the battery pack itself. These lines can be positioned between the individual battery cells and the battery management device to transmit battery data. Therefore, by considering the length, path, and protection devices of the wiring, sensing lines can be incorporated into the battery modules to achieve optimal performance according to the purpose and intended use of the battery system.

[0078] The processor 110 can generate sensing line group data to remove the influence of sensing lines during the diagnostic process, and can perform error correction to remove the influence of sensing lines by correcting the differences between sensing line group data.

[0079] In this case, the processor 110 can derive a correction value at the point where the difference between the sensing line group data is greatest, based on the average of the correction target data included in the multiple sensing line group data and the average of the remaining data.

[0080] The specific details of error correction regarding the influence of the correction sensing line will be described below with reference to Figure 4 and subsequent figures.

[0081] Additionally, the processor 110 can diagnose whether the battery module is abnormal by applying weights determined based on the battery module's structure to the feature values. Here, the weights can refer to values ​​used to correct errors that depend on the battery module's structure and location.

[0082] In other words, the processor 110 can calculate the proximity of multiple battery cells included in the battery module, and apply different weights to the feature values ​​based on the proximity, so that different weights are applied to battery cells that are relatively far apart from each other.

[0083] Accordingly, the processor 110 can correct errors caused by differences in the position of individual battery cells or differences in the sensing structure, rather than errors caused by battery degradation or chemical properties.

[0084] In addition, the processor 110 can generate a trend line obtained by removing outliers based on the correlation between voltage values ​​and feature values ​​included in the battery data, and can diagnose whether the battery pack is abnormal based on the distance from the trend line.

[0085] The following describes embodiments for diagnosing whether a battery pack is abnormal by analyzing the correlation between multiple characteristic values, and may include various other embodiments. For example, processor 110 may analyze the correlation between the degree of battery degradation and characteristic values, or analyze the correlation between internal resistance and characteristic values.

[0086] As described above, the device 1 for diagnosing a battery according to the embodiment can correct errors caused by the shape of the sensing lines included in the battery module, and the accuracy of the diagnosis can be improved by taking into account the structure of the battery module.

[0087] Figure 3 The diagram schematically illustrates a process for determining the state of a battery using a battery diagnostic apparatus according to an embodiment. Figure 3 In this configuration, configurations 101 to 104 are implemented as software blocks and can be stored in memory 120 and executed by processor 110.

[0088] refer to Figure 3 At least one processor 110 can receive battery data acquired by applying a current signal to the battery pack from an externally provided battery data acquisition device 2 or included in the device 1 for diagnosing the battery.

[0089] Here, the current signal applied to the battery module can include the discharge current signal after charging, and the current signal can include a constant current signal (CC signal). Specifically, the current signal applied to the battery module can be applied as a single constant current signal or as a repeated constant current signal when charging and discharging are repeated.

[0090] In other words, the current signal applied to the battery module can be a constant current signal suitable for analysis by dividing its components based on the fact that the voltage change pattern differs for each chemical reaction. In this case, battery data acquisition cycles can be performed at a resolution of 1 mV or lower and measurement cycles of 10 ms or lower, and various modified examples can be included.

[0091] Subsequently, the preprocessing data generation unit 101 of the control unit 100 can generate preprocessing data by removing noise and the influence of sensing lines from the battery data.

[0092] Specifically, the processor 110 can generate preprocessed data by classifying charging, discharging, and resting segments based on the synchronicity between repeated or multiple applied signals. This mitigates the accuracy degradation caused by the voltage resolution of the battery data acquisition device 2 and removes noise.

[0093] Additionally, the processor 110 can generate preprocessed data by removing a preset initial time from a single charge or discharge data. This removes noise caused by the rapid response characteristics of individual battery cells in the current-changing range, as well as noise generated during the individual battery cell sensing process.

[0094] Furthermore, the processor 110 can generate preprocessed data by filtering noise at certain times and frequencies based on various filtering techniques. Accordingly, fundamental and harmonic noise, primarily at 50Hz and 60Hz, can be removed.

[0095] Additionally, the processor 110 can generate preprocessed data by dividing the battery data into structurally identical groups based on the pattern caused by the shape of the sensing line, defining the differences in voltage behavior between the groups, and correcting the differences between the groups by digitizing these differences.

[0096] Subsequently, the linear transformation data generation unit 102 of the control unit 100 can generate linear transformation data obtained by performing a linear transformation on the individual cell voltage data. Here, the linear transformation may include calculations performed on the individual cell voltage, such as calculating the change in the individual cell voltage or calculating the differential value of the individual cell voltage. In the following text, the values ​​included in the linear transformation data may be referred to as linear transformation values.

[0097] The linear transformation data may include linear transformation values ​​calculated based on the individual cell voltage. For example, the linear transformation value may include the change in individual cell voltage and the derivative obtained by differentiating the individual cell voltage based on the time-related change. Specifically, the processor 110 may generate the linear transformation data based on the individual cell voltage data included in the battery data, according to Equation 1 or Equation 2 below.

[0098] [Equation 1]

[0099] (V nt - V t ) / Time change

[0100] [Equation 2]

[0101] (V t2 - V t1 ) / Time change

[0102] Here, the time change can be a time segment corresponding to the difference between two time points (e.g., from the nth time point nt to the first time point t), and can include a time-related segment corresponding to the difference between two time-related values ​​(e.g., from log(nt) to log(t)).

[0103] Processor 110 can generate linear transformation data based on the rate of change of V value in a first time interval (t to nt) (a slope value obtained by performing a first-order linear fit on the V value of the first time interval) and the rate of change of V value in a second time interval (t1 to t2) (a slope value obtained by performing a first-order linear fit on the V value of the second time interval). The reason for generating linear transformation data may be to clearly examine the changes in voltage values ​​included in the battery data (e.g., voltage changes over time), and to more easily determine the trend of the data for individual battery cells by generating linear transformation data. The equations and algorithms used to generate linear transformation data are exemplary, and modified embodiments for generating linear transformation data may be included.

[0104] In addition, the processor 110 can perform normalization on the linear differential value derived based on Equation 1 or Equation 2 above, and can derive eigenvalues ​​from the linear differential value.

[0105] The feature value extraction unit 103 of the control unit 100 can extract and correct feature values ​​by selecting specific time points from the curve obtained by linear transformation, or by extracting and correcting individual components based on equations about chemical reactions.

[0106] In this case, the processor 110 can extract feature values ​​of the time region in which the influence of the sensing line is minimal, and the time region in which the influence of the sensing line is minimal can be obtained during the error correction process performed by the processor 110, or may include a preset time region.

[0107] Examples of feature values ​​extracted by processor 110 may include the maximum linear differential value obtained from the linear differential value curve, the time point at the maximum linear differential value, the linear differential value at a specific time point, and may include various modified examples.

[0108] Subsequently, the battery status diagnostic unit 104 of the processor 110 can diagnose the status of a single battery cell based on the correlation between multiple feature values.

[0109] Specifically, processor 110 can remove outliers that exhibit peculiar behavior in a graph showing the correlation between multiple feature values, and outliers can refer to data points that fall outside the normal range of other data points based on visual or statistical methods.

[0110] The processor 110 can generate a trend line for the group from which outliers have been removed, calculate the distance value of each battery cell from the trend line, and diagnose battery cells whose distance value from the trend line is greater than or equal to a reference value as abnormal battery cells.

[0111] Subsequently, the processor 110 can send information about battery cells diagnosed as abnormal or normal via the communication interface 200 to the external device 3, and thus, the user associated with the battery can remotely check the diagnostic results of the battery cells.

[0112] Figure 4a The figure shows a voltage curve of the battery voltage over time according to an embodiment. Figure 4b The figure shows a linear transformation curve before error correction utilized in a battery diagnostic apparatus according to an embodiment.

[0113] refer to Figure 4a The voltage curve can include the voltage value of a single battery cell over time during the discharge process. Figure 4a The voltage curve can be included in the battery data. (Reference) Figure 4a and Figure 4b The processor 110 can process battery data (e.g., Figure 4a The voltage curve is preprocessed to obtain, for example, Figure 4b The linear transformation curve in the graph.

[0114] refer to Figure 4b The linear transformation curves show that in region (a), it is difficult to determine the trend of the data for individual battery cells, but in region (b), it can be seen that the data is divided into the first group (#1, #3, #5 and #7) and the second group (#2, #4 and #6).

[0115] The linear transformation curve can include the differential value of the battery voltage. Here, the differential value can be the differential value of the voltage based on the time-varying amount of change. The processor 110 can obtain preprocessed data by preprocessing the battery data, and obtain the linear transformation value obtained by performing a linear transformation including noise removal and differentiation based on the preprocessed data. If the linear transformation value of the preprocessed data is generated as a curve, it can be represented as follows: Figure 4b As shown in the image.

[0116] in this case, Figure 4b The horizontal axis can represent time, and the vertical axis can represent the linear transformation value. Additionally, each line can represent a different battery cell, and the index on the right side of the graph can represent the battery cell's number.

[0117] In reference region (a), even when processor 110 derives the logarithmic derivative from the battery module under the same conditions, it can be seen that the linear transformation data at t0 and t... B They are divided into two groups. Here, t0 and t B It can correspond to the voltage measurement time point (units are seconds or milliseconds, etc.).

[0118] Such separation may occur due to the influence of sensing lines that sense battery data, and may also occur, for example, due to mutual coupling where two or more electrical components interact with each other’s electromagnetic fields and influence each other in the opposite direction to the intended effect.

[0119] Specifically, the impact of separation phenomena on battery diagnostics will be related to... Figure 4b Area (b) and Figure 5 as well as Figure 4b Region (c) and Figure 6 Let's describe them together.

[0120] Figure 5 The diagram illustrates the characteristic values ​​used in a battery diagnostic apparatus according to an embodiment before error correction, and... Figure 6Other characteristic values ​​prior to error correction are shown in the apparatus for diagnosing batteries according to an embodiment.

[0121] First, let's refer to each other. Figure 4b Area (b) and Figure 5 ,exist Figure 4b In region (b), due to the influence of the sensing line, the logarithmic differential curve of each battery cell can be displayed as being divided into cells 4, 6, and the rest (cells 1, 2, 3, 5, and 7). Thus, if from... Figure 4b By extracting feature values ​​from region (b), feature values ​​in a jagged pattern in the vertical direction can be derived, such as... Figure 5 As shown in the image.

[0122] Continue to refer to Figure 5 The indices m1 to m4 can refer to the battery module numbers. Additionally, the horizontal axis can represent the individual cell voltages in the resting state before the signal is applied, and the vertical axis can represent the voltages from... Figure 4b The feature values ​​extracted from region (b).

[0123] When the influence of the sensing line is not corrected, there is a possibility that the location of the characteristic values ​​of the individual units included in the specific module to be diagnosed may hinder accurate diagnosis. For example, when there are units with characteristics corresponding to those in horizontal or vertical positions... Figure 5 When the values ​​of abnormal monomers with similar characteristics to monomers 28, 26, 24 and 22 corresponding to module 4 shown in the figure are similar, it may hinder accurate diagnosis.

[0124] Similarly, refer to them together. Figure 4b Region (c) and Figure 6 Compared to region (b), in Figure 4b The influence of the sensing line is reduced in region (c), allowing the differences in the linear transformation curves for each battery cell to be reduced and displayed. Thus, when from... Figure 4b When extracting feature values ​​from region (c), feature values ​​with a jagged shape in the vertical direction can be derived, such as... Figure 6 As shown, it is in the following form: with Figure 5 In comparison, the differences between eigenvalues ​​are reduced.

[0125] Continue to refer to Figure 6 The indices m1 to m4 can refer to the battery module number, similar to... Figure 5 Additionally, the horizontal axis can represent the unit voltage in the resting state before the signal is applied, and the vertical axis can represent the voltage from... Figure 4b The feature values ​​extracted from region (c).

[0126] When the influence of the sensing line is reduced, the characteristic values ​​of individual cells can be distinguished from each other, thus improving the accuracy of diagnosis.

[0127] Based on this fact, the device 1 for diagnosing batteries according to the embodiment can correct for the influence of the sensing line, and the specific correction principle and results will be referred to Figure 7 and Figure 8 Describe it.

[0128] Figure 7 The structure of a battery pack is shown, which is the diagnostic target of a device for diagnosing batteries according to an embodiment. Figure 8 The diagram shows a linear transformation curve after error correction, utilized in a battery diagnostic apparatus according to an embodiment.

[0129] refer to Figure 7 An example of the structure of the battery pack P, battery modules M-1, M-2, M-3 and M-4, and battery cells #1 to #14 performed on the battery in the device 1 for diagnosing the battery according to an embodiment is shown.

[0130] Specifically, individual battery cells #1 to #14 can constitute battery modules M-1 and M-2, and multiple battery modules M-1, M-2, M-3, and M-4 can constitute battery pack P. Each of the sensing lines a-1, a-2, b-1, and b-2 monitors the state of each individual cell, and power lines c-1 and c-2 can transmit power to battery pack P.

[0131] In the case of battery cells #1 to #7 included in battery module M-1, sensing can be performed for odd-numbered battery cells #1, #3, #5 and #7 and even-numbered battery cells #2, #4, #6 and #8 through different sensing lines.

[0132] In addition, since the battery cells #1 to #7 included in battery module M-1 are affected by different power lines c-1 and c-2 due to their different physical locations and structures, differences may occur even in battery data obtained under the same conditions.

[0133] Furthermore, each of the battery modules M-1, M-2, M-3, and M-4 may have different heat distributions depending on its location, and the pressure applied to it may also vary depending on the stacking method and structure. Therefore, the voltage, internal resistance, and capacity of each of the battery modules M-1, M-2, M-3, and M-4 may vary, and the processor 110 can apply weights to the data of each battery module according to its location to correct for the differences.

[0134] Specifically, the processor 110 can calculate the proximity of multiple battery cells included in the battery module and apply different weights based on the proximity, and the processor 110 can apply different weights to the feature values ​​based on the relative positions of the power lines connected to the multiple battery modules and battery packs.

[0135] In other words, since the feature values ​​extracted from nearby battery cells within the battery module exhibit similar behavior, the processor 110 assigns them weights with the same or small differences, and since the feature values ​​extracted from battery cells far from the battery module exhibit different behavior, the processor 110 can correct for differences in behavior by assigning them weights with different or large differences.

[0136] Additionally, the processor 110 can set weights when designing the battery system to correct for electromagnetic effects on multiple sensing lines or power lines, such as... Figure 7 As shown in the figure, the optimal weights can be derived using algorithms or machine learning models.

[0137] According to an embodiment, the device 1 for diagnosing batteries can eliminate the influence of the sensing line through equivalent analysis to correct the influence of the sensing line.

[0138] Specifically, the direction or path of the sensing loop differs between even-numbered and odd-numbered battery cells (e.g.) Figure 7 In the case shown in the exemplary structure, the mutual coupling modes may occur in opposite directions, and therefore the individual cells exhibit opposite trends for a certain period of time immediately following the change in current.

[0139] In an embodiment, as one of the methods for correcting the influence of the sensing line, the processor 110 may perform correction for odd-numbered battery cells based on the following: i) linear transformation values ​​measured from existing odd-numbered battery cells, ii) the average value of the linear transformation values ​​in odd-numbered battery cells (first average value), iii) the average value of the linear transformation values ​​in even-numbered battery cells (second average value), and iv) the value based on the average value of the linear transformation values ​​in odd-numbered battery cells (first average value) and the average value of the linear transformation values ​​in even-numbered battery cells (second average value) (e.g., the arithmetic mean of the first average value and the second average value).

[0140] The reason for calculating the first and second average values ​​separately could be that the number of odd-numbered battery cells can differ from the number of even-numbered battery cells. When calculating the average of the linear transformation values ​​across all battery cells without distinguishing between odd and even-numbered cells, the proportion associated with the linear transformation values ​​in the odd-numbered cells may be reflected more significantly. For example, when there are four odd-numbered battery cells with linear transformation values ​​of {1, 3, 8, 12}, and three even-numbered battery cells with linear transformation values ​​of {2, 4, 6}, the first average value can be calculated as 6 (corresponding to value ii above), the second average value can be calculated as 4 (corresponding to value iii above), and the average of the first and second average values ​​can be calculated as 5 (corresponding to value iv above). On the other hand, if the average of all linear transformation values ​​is calculated without distinguishing between odd and even-numbered battery cells, the average value can be calculated as 5.14, which is closer to the first average value. In summary, if the average value of the linear transformation of all battery cells is calculated, the linear transformation values ​​of odd-numbered battery cells may be more significantly reflected in the average value. Therefore, the influence of the sensing line can be corrected more accurately by distinguishing between the first average value and the second average value and calculating the first average value and the second average value.

[0141] Specifically, processor 110 can obtain the corrected linear transformation value for odd-numbered battery cells by deriving the value obtained by subtracting iv) from ii) as the correction value v) and excluding the correction value v) from i). The aforementioned linear transformation value may include the value obtained by differentiating the voltage value V of each battery cell based on the amount of change over logarithmic time (e.g., log(t)).

[0142] Similarly, the processor 110 may perform corrections for even-numbered battery cells based on the following to correct for the effects of the sensing line: i) linear transformation values ​​measured from existing even-numbered battery cells; ii) the average value of the linear transformation values ​​in even-numbered battery cells; and iii) the value based on the average value of the linear transformation values ​​in even-numbered battery cells (second average value) and the average value of the linear transformation values ​​in odd-numbered battery cells (first average value) (e.g., the arithmetic mean of the first average value and the second average value).

[0143] Specifically, the processor 110 can obtain the corrected linear transformation value of the even-numbered battery cells by deriving the value obtained by subtracting iii) from ii) as the correction value iv) and excluding the correction value iv) from i).

[0144] However, Figure 7The battery structure is exemplary, and various structural modifications may exist depending on the battery shape (e.g., square, cylindrical, pouch, etc.) and the application product of the battery (e.g., electric vehicles, ESS, robots, etc.). Therefore, various examples of modifications may exist in the method for correcting the influence of the sensing line (e.g., methods using electromagnetic models, methods using scaling adjustments, etc.), and correction can be performed in a manner equivalent to that of the device 1 for diagnosing the battery according to the embodiment through equivalent analysis.

[0145] In this way, the processor 110 can obtain, for example, by removing the influence of the sensing line. Figure 8 The linear transformation curve shown is illustrated in the figure.

[0146] like Figure 8 As shown, processor 110 can remove noise from battery data, perform error correction to obtain preprocessed data, and perform a linear transformation on the preprocessed data to obtain linear transformation values. If the linear transformation values ​​of the preprocessed data, after correcting for the effects of the sensing lines, are generated as a graph, the linear transformation values ​​can be represented as follows: Figure 8 As shown in the image.

[0147] in this case, Figure 8 The horizontal axis can represent time (e.g., the unit could be [ms]), and the vertical axis can represent the linear transformation value. Additionally, each line can represent a different battery cell, and the index on the right side of the graph can represent the battery cell number.

[0148] In other words, Figure 4b In this context, because the influence of the sensing line is not corrected, the curves corresponding to individual battery cells may be represented by dividing the data into two groups. When extracting feature values ​​from the data divided into two groups, inaccurate battery diagnoses may be made.

[0149] Therefore, the device 1 for diagnosing a battery according to the embodiment can perform error correction to remove the influence of the sensing line in order to mitigate inaccurate battery diagnosis. If the processor 110 executes as follows... Figure 8 The error correction shown in the figure forms a group of linear transformation values ​​for each battery cell, so that a uniform characteristic value can be derived based on the height of the linear transformation values.

[0150] In the following, examples of deriving the feature values ​​used in the device 1 for diagnosing batteries according to an embodiment will be described.

[0151] Specifically, the processor 110 can select meaningful time points in the linear transformation curve, extract the value at that time point, and use that value as a feature value. Furthermore, through linear transformation, the chemical reactions of the battery can be analyzed based on equations and extracted as individual components.

[0152] In this process, the processor 110 can select the feature values ​​of the time region least affected by the sensing line, and the process of selecting the feature values ​​of the time region least affected by the sensing line can be executed in parallel with the task of correcting the influence of the sensing line.

[0153] Figure 9 The characteristic values ​​included in the discharge curve diagram according to an embodiment are shown. In this case, Figure 9 The horizontal axis can represent time (e.g., the unit could be [ms]), and the vertical axis can represent the voltage of a single unit.

[0154] Additionally, the indices on the right can refer to different battery cells #1 to #7, and each line can refer to the discharge curve of each of battery cells #1 to #7.

[0155] Specifically, processor 110 can extract multiple different feature values ​​by cropping specific segments of the discharge curve. That is, processor 110 can determine the resistance value at a specific time point (e.g., 10ms) as feature 1 (f_1) based on Equation 2 below.

[0156] [Equation 3]

[0157] Additionally, the processor 110 can determine the slope of a first-order linear fit for a specific segment (e.g., the 900 to 1000 ms segment) as feature 2 (f_2).

[0158] To this end, processor 110 can perform first-order linear fitting based on techniques such as least squares, maximum likelihood estimation, and minimum absolute deviation, and processor 110 can find a straight line of the form y=mx+b for time t and voltage V.

[0159] Here, y can refer to voltage V, which is the dependent variable, x can refer to time t, which is the independent variable, m can refer to the slope corresponding to feature 2 (f_2), and b can refer to the y-intercept.

[0160] Additionally, the processor 110 can determine the resistance value at a specific time point (e.g., 1000ms) as feature 3 (f_3) based on Equation 4 below.

[0161] [Equation 4]

[0162] Specifically, the processor 110 can generate diagnostic indicators by correlation analysis with other feature values ​​similar to feature 0 (f_0), feature 1 (f_1), feature 2 (f_2) and feature 3 (f_3), and can diagnose the battery by utilizing the generated diagnostic indicators.

[0163] Figure 10 The diagram illustrates feature values ​​included in a linear transformation curve used in a battery diagnostic apparatus according to an embodiment. In this case, Figure 10 The horizontal axis can represent time (e.g., the unit could be [ms]), and the vertical axis can represent the linear transformation value.

[0164] Additionally, the indices on the right can refer to different battery cells #1 to #7, and each line can refer to a linear transformation curve of each of battery cells #1 to #7.

[0165] Processor 110 can extract multiple distinct feature values ​​from a graph obtained after removing the initial segment (e.g., the initial 3ms) of the linear transformation graph and applying a filter with a cutoff frequency of 60Hz using a low-pass filter (LPF).

[0166] Specifically, the processor 110 can determine the linear transformation value with the maximum value as feature 4 (f_4), the linear transformation value at any time point (e.g., 100ms) as feature 5 (f_5), and the final point of the battery cell #7 whose highest point has the minimum value as feature 6 (f_6).

[0167] Additionally, the processor 110 can be used at any point in time (e.g., t E The linear transformation value at 490 ms is determined as feature 7 (f_7), and the time point [ms] where the linear transformation value has the maximum value is determined as feature 8 (f_8).

[0168] Figure 11 The illustration shows feature values ​​included in an average curve graph used in a battery diagnostic apparatus according to an embodiment. In this case, Figure 11 The horizontal axis can represent time (e.g., the unit could be [ms]), and the vertical axis can represent the linear transformation value.

[0169] The processor 110 can obtain linear transformation values ​​from all the individual cells included in the battery module, and extract multiple feature values ​​by partially cropping the curve obtained by averaging the linear transformation values ​​of the entire battery module.

[0170] The processor 110 can derive the maximum and minimum values ​​of the linear transformation value from a graph obtained by averaging the linear transformation value of the entire battery module, and can determine the time at the corresponding maximum and minimum values ​​as T. max and T min .

[0171] Subsequently, processor 110 can use the following equations 5 to 7 to derive feature 9 (f_9) and feature 10 (f_10).

[0172] [Equation 5]

[0173] [Equation 6]

[0174] [Equation 7]

[0175] Processor 110 can T max To T min The time point at the 1 / 4 division of the segment is determined as f_10s, and T is... max To T min The time point at the 3 / 4 division of the segment is determined as f_10e. In this case, if f_10s is a value from 20ms ago, then f_10s can be fixed at 20ms.

[0176] After that, processor 110 can time point T mid The linear transformation value is determined as feature 9 (f_9), and the slope after first-order linear fitting of the segment from f_10s to f_10e can be determined as f_10.

[0177] Thus, the device 1 for diagnosing batteries according to the embodiment can select feature values ​​from various battery data and linear transformation curves, and diagnose the battery based on the correlation of the feature values, enabling rapid battery inspection for various environments.

[0178] Figures 12 to 14 The characteristic values ​​derived from the device for diagnosing batteries according to an embodiment are shown.

[0179] First, refer to Figure 12 The vertical axis represents the feature value A extracted from a relatively short time period within the time segment of the acquired battery data, and the horizontal axis represents the feature value B extracted from a relatively long time period within the time segment of the acquired battery data.

[0180] Additionally, the index m1 on the right side of the graph can refer to the battery module number.

[0181] because Figure 12 The graph shown in the middle is the result after correcting for the influence of the sensing line, so the points corresponding to the characteristic values ​​A and B of the battery cells indicated by the solid circles can be identified. Figure 12 The points in the diagram can correspond to the characteristic values ​​A and B of a normal battery cell. Although in Figure 12 Although not shown in the diagram, the points corresponding to the characteristic values ​​A and B of the abnormal battery cells can be clearly distinguished from the points corresponding to the normal cells in the horizontal and / or vertical positions.

[0182] The device 1 for diagnosing batteries according to the embodiment, even when Figure 12 It can also diagnose whether the battery is abnormal under certain conditions, and can improve the accuracy of diagnosis by removing outliers with specific behaviors based on battery characteristics.

[0183] In other words, line (a) can refer to the trend line of a normal battery cell, and processor 110 can identify battery cells that are separated from line (a) by a preset distance or greater as abnormal battery cells, and identify battery cells detected at a distance less than the preset distance as normal battery cells.

[0184] Here, statistical techniques such as mean, standard deviation, interquartile range, etc., or machine learning techniques such as isolated forest, local outlier factor (LOF), one-class support vector machine (one-class SVM), etc., can be used as methods for processor 110 to remove outliers.

[0185] Thus, the device 1 for diagnosing batteries according to the embodiment can identify trends in the normal range within a group in which outliers with unusual behavior have been removed from all battery cells, and thus has the effect of improving the accuracy of the analysis.

[0186] Next, refer to Figure 13 The vertical axis represents the feature value C extracted from a relatively short time period within the time segment of the acquired battery data, and the vertical axis represents the feature value D extracted from a relatively long time period within the time segment of the acquired battery data.

[0187] Additionally, the index m2 on the right side of the graph can refer to the battery module number.

[0188] because Figure 13 The graph in the middle figure shows the result after correcting for the influence of the sensing line, so the points indicated by solid circles can be points corresponding to the characteristic values ​​C and D of a normal battery cell. Although in Figure 13 Although not shown in the diagram, the points corresponding to the characteristic values ​​C and D of the abnormal battery cells can be clearly distinguished from the points corresponding to the normal cells in the horizontal and / or vertical positions.

[0189] and Figure 12 Similarly, line (a) can refer to the trend line of a normal battery cell, and processor 110 can identify battery cells that are separated from line (a) by a preset distance or greater as abnormal battery cells, and identify battery cells detected at a distance less than the preset distance as normal battery cells.

[0190] Next, refer to Figure 14 The vertical axis represents the feature value E extracted from a relatively short time period within the time segment of the acquired battery data, and the vertical axis represents the feature value F extracted from a relatively long time period within the time segment of the acquired battery data.

[0191] Additionally, the index m3 on the right side of the graph can refer to the battery module number.

[0192] because Figure 14 The graph in the middle figure shows the result after correcting for the influence of the sensing line, so the points indicated by solid circles can be points corresponding to the characteristic values ​​E and F of a normal battery. Although in Figure 14 Although not illustrated, the points corresponding to the characteristic values ​​E and F of the abnormal cells can be clearly distinguished from the points corresponding to normal cells in the horizontal and / or vertical positions.

[0193] and Figure 12 Similarly, line (a) can refer to the trend line of a normal battery cell, and processor 110 can identify battery cells that are separated from line (a) by a preset distance or greater as abnormal battery cells, and identify batteries detected at a distance less than the preset distance as normal battery cells.

[0194] In this way, the device 1 for diagnosing the battery can extract multiple feature values, derive the correlation between the feature values, and separately represent normal and abnormal battery cells. Therefore, the optimal feature values ​​for battery diagnosis can be derived to increase the efficiency of battery diagnosis.

[0195] Figure 15 A control flowchart for a method for diagnosing a battery according to an embodiment is shown, and Figure 16 Continue to show in Figure 15 The following is a control flowchart of a method for diagnosing a battery according to an embodiment.

[0196] refer to Figure 15The processor 110 can receive battery data (1500) acquired by applying a current signal to the battery module. Here, the battery data acquired by applying a current signal may include battery data acquired by the processor 110 applying a current signal to the battery module, battery data acquired by the external device 3, and battery data received by the server device.

[0197] Subsequently, the processor 110 can generate multiple sensing line group data (1510) according to the sensing line shape of the battery module, and the processor 110 can determine whether the difference between the sensing line group data exceeds a preset value (1520).

[0198] In other words, when it is determined that the difference between the sensing line group data does not exceed a preset value (no at 1520), the processor 110 may not perform error correction to remove the influence of the sensing lines, because the influence of the sensing lines is canceled out and does not affect feature value extraction.

[0199] On the other hand, when it is determined that the difference between the sensing line group data exceeds a preset value (yes at 1520), the processor 110 can derive a correction value (1530) based on the average value of the correction target data and the average value of the remaining data.

[0200] Subsequently, the processor 110 can use the correction value to correct the differences between the sensing line group data (1540). Here, the processor 110 can correct the sensing line effect by eliminating the differences between the sensing line group data through equivalent analysis.

[0201] Continue to refer to Figure 16 The processor 110 can calculate the proximity of each battery cell (1600) based on multiple reference battery cells. In other words, the processor 110 can derive the proximity based on the position of each battery cell.

[0202] Subsequently, the processor 110 can determine whether the proximity is less than a reference value (1610) based on multiple reference battery cells. That is, the processor can determine the proximity of the corresponding battery cell to be located based on reference battery cells that can define the relative positions of the battery cells.

[0203] If the proximity to a reference value is determined to be less than the reference value based on multiple reference battery cells (Yes at 1610), then processor 110 can apply the same weight to the battery cells (1620). That is, processor 110 can minimize environmental errors in battery diagnostics by applying the same weight to adjacent battery cells based on the fact that adjacent battery cells are exposed to similar environments (e.g., temperature, pressure, etc.).

[0204] Accordingly, a unified battery diagnostic environment can be created by correcting for distortions in battery data caused by physical characteristics such as the location and structure of the battery module, as well as distortions in battery data caused by the sensing structure.

[0205] Subsequently, the processor 110 can generate a trend line (1630) obtained by excluding outliers based on the correlation between voltage values ​​and feature values ​​included in the battery data, and the processor 110 can derive the distance value between the battery cell and the trend line (1640).

[0206] Here, the distance value between the battery cell and the trend line can refer to the distance value between the location of the feature value measured from the battery cell and the trend line, and the processor 110 can derive the distance value using distance measurement algorithms such as Euclidean distance and Dijkstra's algorithm.

[0207] Subsequently, the processor 110 can diagnose whether a battery cell is abnormal based on the distance value. In an embodiment, the processor 110 can determine whether the distance value between the derived battery cell and the trend line is greater than or equal to a preset reference value. If it is determined that the distance value between the derived battery cell and the trend line is less than the preset reference value, the processor 110 can diagnose the battery cell as normal; and if it is determined that the distance value between the derived battery cell and the trend line exceeds the preset reference value, the processor 110 can diagnose the battery cell as abnormal.

[0208] Thus, the device 1 for diagnosing batteries according to the embodiment can remove the influence of the battery module structure when diagnosing abnormalities such as lithium plating in individual battery cells, and can accurately and quickly diagnose batteries based on various characteristic values, thereby shortening the time required for battery diagnosis and reducing costs.

[0209] Furthermore, the disclosed embodiments can be implemented in the form of a recording medium storing computer-executable instructions. These instructions can be stored as program code, which, when executed by a processor, can generate program modules to perform the operations of the disclosed embodiments. The recording medium can be implemented as a computer-readable recording medium.

[0210] Computer-readable recording media include all types of recording media that store instructions that can be decoded by a computer. For example, computer-readable recording media may include, for example, read-only memory (ROM), random access memory (RAM), magnetic tape, magnetic disk, flash memory, optical data storage devices, etc.

[0211] Additionally, computer-readable recording media may be provided in the form of non-transitory storage media. Here, the term "non-transitory storage media" refers only to a tangible device that does not contain signals (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently in a storage medium and cases where data is temporarily stored. For example, a "non-transitory storage medium" may include a buffer for temporarily storing data.

[0212] According to embodiments, methods according to the various embodiments disclosed in this document can be provided by being included in a computer program product. This 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 (e.g., the Play Store™), directly between two user devices (e.g., smartphones), or online (e.g., downloaded or uploaded). In the case of online distribution, at least a portion of the computer program product (e.g., a downloadable application) can be temporarily stored or temporarily generated in a machine-readable recording medium, such as the memory of a manufacturer's server, an app store's server, or an intermediate server.

[0213] In the foregoing, although all components constituting the embodiments disclosed in this document have been described as being combined or operated in combination, the embodiments disclosed in this document are not necessarily limited to these 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 combination with one or more of them.

[0214] Unless otherwise specifically stated to the contrary, the terms "comprising," "configured," or "having" above mean that the corresponding component may be included therein, and therefore should be interpreted as 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. Common terms, such as those defined in dictionaries, should be interpreted according to the context of the relevant art, and should not be interpreted in an idealized or overly formal sense unless explicitly defined in this document.

[0215] The above description is merely an illustrative description of the technical concepts disclosed in this document, and those skilled in the art to which the embodiments disclosed in this document pertain can make various modifications and variations without departing from the basic characteristics of the embodiments disclosed in this document. Therefore, the embodiments disclosed in this document are not intended to limit the technical concepts of the embodiments disclosed in this document, but rather for descriptive purposes, and the scope of the technical concepts disclosed in this document is not limited by these embodiments. The scope of protection of the technical concepts disclosed in this document should be interpreted by the scope of the appended claims, and all technical concepts within the equivalent scope should be interpreted as being included within the scope of the rights in this document.

Claims

1. A device for diagnosing a battery, comprising: A communication interface for receiving battery data acquired by applying a current signal to the battery module; as well as At least one processor is configured to perform error correction to correct for the influence of sensing lines related to the structure of the battery module and generate preprocessed data based on the battery data, extract feature values ​​based on the preprocessed data, and diagnose the state of the battery module based on multiple feature values ​​extracted at different time regions.

2. The apparatus of claim 1, wherein, The at least one processor applies a linear transformation to the preprocessed data to obtain linearly transformed data, and extracts the feature values ​​based on the linearly transformed data.

3. The apparatus of claim 1, wherein, The at least one processor groups the battery data according to the shape of the sensing line of the battery module to generate multiple sensing line group data, and generates the preprocessed data by correcting the differences between the multiple sensing line group data using a correction value.

4. The apparatus according to claim 3, wherein, The at least one processor derives the correction value at the point where the difference between the multiple sensing line group data is the largest, and uses the correction value to correct the difference between the multiple sensing line group data.

5. The apparatus according to claim 4, wherein, The at least one processor derives the correction value based on the average of the correction target data included in the plurality of sensing line group data and the average of the remaining data.

6. The apparatus according to claim 5, wherein, The at least one processor derives the correction value as the value obtained by excluding the average of the average of the correction target data and the average of the remaining data from the average of the correction target data, and uses the correction value to correct the difference.

7. The apparatus according to claim 1, wherein, The at least one processor diagnoses the state of the battery module by applying weights determined based on the structure of the battery module to the plurality of feature values.

8. The apparatus according to claim 7, wherein, The at least one processor calculates the proximity of the plurality of battery cells included in the battery module, and diagnoses the state of the battery module by applying the weights to the plurality of feature values ​​differently based on the proximity.

9. The apparatus according to claim 8, wherein, The at least one processor diagnoses the state of the battery module by applying the weights differently to the plurality of feature values ​​based on the relative positions of the power lines connected to the plurality of battery cells and the battery module.

10. The apparatus according to claim 1, wherein, The at least one processor generates a trend line obtained by removing outliers based on the correlation between multiple feature values ​​derived from differences in chemical reactions occurring over time, and diagnoses the state of the battery module based on the trend line.

11. The apparatus according to claim 10, wherein, The at least one processor derives the distance value between each battery cell included in the battery module and the trend line, and diagnoses battery cells whose distance value is greater than or equal to a preset reference value as abnormal.

12. A method for diagnosing a battery, comprising: Receive battery data acquired by applying a current signal to the battery module; Perform error correction to correct the influence of sensing lines related to the structure of the battery module and generate preprocessed data based on the battery data; Feature values ​​are extracted based on the preprocessed data; as well as The state of the battery module is diagnosed based on multiple feature values ​​extracted at different time zones.

13. The method according to claim 12, wherein, Extracting the feature values ​​involves applying a linear transformation to the preprocessed data to obtain linearly transformed data, and then extracting the feature values ​​based on the linearly transformed data.

14. The method according to claim 12, wherein, Generating the preprocessed data includes grouping the battery data according to the shape of the sensing line of the battery module to generate multiple sensing line groups of data, and generating the preprocessed data by correcting the differences between the multiple sensing line groups of data using a correction value.

15. The method according to claim 14, wherein, Correcting the differences between the multiple sensing line group data includes deriving the correction value at the point where the difference value between the multiple sensing line group data is the largest, and using the correction value to correct the differences between the multiple sensing line group data.

16. The method according to claim 15, wherein, The correction value is derived by deriving the correction value based on the average of the correction target data included in the plurality of sensing line group data and the average of the remaining data.

17. The method according to claim 16, wherein, Deriving the correction value involves deriving the correction value as the average obtained by excluding the average of the average of the correction target data and the average of the remaining data from the average of the correction target data.

18. The method according to claim 12, wherein, Diagnosing the state of the battery module includes applying weights determined based on the structure of the battery module to the plurality of feature values.

19. The method according to claim 18, wherein, Diagnosing the state of the battery module includes calculating the proximity of multiple battery cells included in the battery module, and diagnosing the state of the battery module by applying the weights to the multiple feature values ​​differently based on the proximity.

20. The method according to claim 19, wherein, Diagnosing the state of the battery module includes applying weights differently to the plurality of feature values ​​based on the relative positions of the power lines connected to the plurality of battery cells and the battery module.

21. The method according to claim 12, wherein, Diagnosing the state of the battery module includes generating a trend line obtained by removing outliers based on the correlation between multiple feature values ​​derived from differences in chemical reactions occurring over time, and diagnosing the state of the battery module based on the trend line.

22. The method according to claim 21, wherein, Diagnosing the state of the battery module includes deriving the distance value between each battery cell in the battery module and the trend line, and diagnosing battery cells whose distance value is greater than or equal to a preset reference value as abnormal.

Images