Battery diagnosis apparatus and method

Abstract

According to some embodiments disclosed in the present document, a battery diagnosis apparatus may comprise: a sensor for measuring battery test data from a battery to be diagnosed; and a processor for dividing the battery test data into a plurality of charging / discharging state datasets on the basis of a plurality of reference values related to a plurality of charging / discharging states, performing preliminary fault diagnosis on each of the plurality of charging / discharging state datasets on the basis of a preliminary diagnosis model, and, on the basis of the result of the preliminary fault diagnosis, making a final determination that the battery to be diagnosed has a fault.

Description

Battery diagnostic device and method

[0001] Cross-citation with related applications

[0002] This application claims the benefit of priority to Republic of Korea Patent Application No. 10-2023-0076815, filed June 15, 2023, the entire contents of which are incorporated herein by reference.

[0003] Technology field

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

[0005] Research and development on secondary batteries has been actively underway recently. The term "secondary battery" refers to a rechargeable battery, encompassing both conventional Ni / Cd and Ni / MH batteries, as well as more recent lithium-ion batteries. Among secondary batteries, lithium-ion batteries boast a significantly higher energy density than conventional Ni / Cd and Ni / MH batteries. Furthermore, lithium-ion batteries can be manufactured in a compact and lightweight form, making them ideal power sources for mobile devices. Recently, their use has expanded to include power sources for electric vehicles, drawing attention as a next-generation energy storage medium.

[0006] Various diagnostic models can be utilized to diagnose whether battery cells, modules, or packs are functioning normally. For example, while a charge / discharge test voltage is applied, the test voltage can be measured from the battery under diagnosis, and a diagnostic model can be applied to this voltage to diagnose whether the battery is defective. These diagnostic models can be used to evaluate model performance using indicators such as the diagnosis rate (selectivity), which indicates the rate at which they correctly diagnose defects, and the over-diagnosis rate, which indicates the rate at which they incorrectly diagnose non-defective batteries as defective. However, typical diagnostic models have a selectivity of less than 80% and an over-diagnosis rate of more than 10%, and their performance cannot be guaranteed for atypical charge / discharge patterns.

[0007] One purpose of the embodiments disclosed in this document is to provide a battery diagnostic device and method having high usability even for atypical charge / discharge patterns.

[0008] The technical objectives of the embodiments disclosed in this document are not limited to the technical tasks mentioned above, and other technical tasks not mentioned will be clearly understood by those skilled in the art from the descriptions below.

[0009] According to some embodiments disclosed in the present document, a battery diagnosis device may include a sensor configured to measure battery test data from a battery to be diagnosed; and a processor configured to divide the battery test data into a plurality of charge / discharge state data sets based on a plurality of reference values ​​associated with a plurality of charge / discharge states, perform a preliminary failure diagnosis on each of the plurality of charge / discharge state data sets based on a preliminary diagnosis model, and determine the battery to be diagnosed as a final failure based on a result of the preliminary failure diagnosis.

[0010] According to some embodiments, the processor may be configured to determine the battery to be diagnosed as a final failure if two or more of the plurality of charge / discharge state data sets are determined to be preliminary failures when determining the final failure.

[0011] According to some embodiments, the processor may be further configured to compare the result of determining the battery to be diagnosed as being defective with the actual defect of the battery to be diagnosed to calculate a diagnosis rate and an over-inspection rate.

[0012] According to some embodiments, the plurality of charge / discharge states may include a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

[0013] According to some embodiments, the processor may be further configured to select, from among combinations of two or more states selected from the plurality of charge / discharge states, a combination that satisfies at least one of a condition in which the diagnostic rate is greater than or equal to a first threshold value and a condition in which the over-inspection rate is less than or equal to a second threshold value.

[0014] According to some embodiments, the battery test data may include at least one of voltage data, current data, and temperature data of the battery to be diagnosed measured during a time period of a charge / discharge test.

[0015] In some embodiments, the preliminary diagnostic model may include at least one of a support vector classifier (SVC) model, an Extra-Trees model, a CatBoost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

[0016] According to some embodiments disclosed in the present document, a battery diagnosis method may include: measuring battery test data from a battery to be diagnosed through a sensor; dividing the battery test data into a plurality of charge / discharge state data sets through a processor based on a plurality of reference values ​​associated with a plurality of charge / discharge states; performing a preliminary failure diagnosis through the processor for each of the plurality of charge / discharge state data sets based on a preliminary diagnosis model; and determining, through the processor, a final failure of the battery to be diagnosed based on a result of the preliminary failure diagnosis.

[0017] According to some embodiments, the step of determining the final failure may include a step of determining the battery to be diagnosed as the final failure if two or more data sets among the plurality of charge / discharge state data sets are determined to be preliminary failures.

[0018] According to some embodiments, the battery diagnosis method may further include a step of comparing the result of determining the battery to be diagnosed as being defective with the actual defect of the battery to be diagnosed to calculate a diagnosis rate and an over-inspection rate.

[0019] According to some embodiments, the plurality of charge / discharge states may include a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

[0020] According to some embodiments, the battery diagnosis method may further include a step of selecting, from among combinations composed of two or more states selected from the plurality of charge / discharge states, a combination that satisfies at least one of a condition in which the diagnosis rate is equal to or greater than a first threshold value and a condition in which the over-inspection rate is equal to or less than a second threshold value.

[0021] According to some embodiments, the battery test data may include at least one of voltage data, current data, and temperature data of the battery to be diagnosed measured during a time period of a charge / discharge test.

[0022] In some embodiments, the preliminary diagnostic model may include at least one of a support vector classifier (SVC) model, an Extra-Trees model, a CatBoost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

[0023] According to the embodiments disclosed in this document, an algorithm can be provided that divides battery test data into a plurality of charge / discharge states and finally determines a battery to be diagnosed as defective when two or more data sets among the plurality of charge / discharge state data sets corresponding to the plurality of charge / discharge states are determined to be defective, so that battery defects can be diagnosed with high performance without being limited to a specific charge / discharge pattern.

[0024] The technical effects according to the embodiments disclosed in this document are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art according to the disclosure of this document.

[0025] FIG. 1 is a drawing illustrating elements constituting a battery diagnostic device according to some embodiments disclosed in this document.

[0026] FIG. 2 is a diagram illustrating battery test data according to some embodiments disclosed in this document.

[0027] FIG. 3 is a diagram illustrating a plurality of reference values ​​representing a plurality of charge / discharge states according to some embodiments disclosed in this document.

[0028] FIG. 4 is a drawing illustrating a structure in which a charge state (Charge) is divided into a rapid charge state (QC) and a constant current charge state (CC) according to some embodiments disclosed in this document.

[0029] FIG. 5 is a diagram illustrating multiple charge / discharge state datasets according to some embodiments disclosed in this document.

[0030] FIG. 6 is a diagram illustrating model performance of preliminary diagnostic models according to some embodiments disclosed in this document for each of a plurality of charge / discharge states.

[0031] FIG. 7 is a diagram illustrating model performance of an ensemble algorithm according to some embodiments disclosed in this document.

[0032] FIG. 8 is a diagram illustrating steps of a battery diagnosis method according to some embodiments disclosed in this document.

[0033] Hereinafter, embodiments described in this document are described with reference to the attached drawings. However, this is not intended to limit the disclosure of this document to specific embodiments, and it should be understood that various modifications, equivalents, and / or alternatives of the embodiments described in this document are included.

[0034] The embodiments and terminology used in this document are not intended to limit the technical features described in this document to a specific embodiment, but should be understood to encompass various modifications, equivalents, or alternatives of the embodiment. In connection with the description of the drawings, similar reference numerals may be used to refer to similar or related components. The singular form of a noun corresponding to an item may include one or more of the item, unless the context clearly indicates otherwise.

[0035] In this document, 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 each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first", "second", "first", "second", "A", "B", "(a)", or "(b)" may be used merely to distinguish the corresponding component from other corresponding components, and do not limit the corresponding components in any other respect (e.g., importance or order) unless specifically stated otherwise.

[0036] In this document, when a component (e.g., a first component) is referred to as being “connected,” “coupled,” or “connected,” with or without the terms “functionally” or “communicatively,” or “coupled” or “connected,” it means that the component can be connected to the other component directly (e.g., wired or wirelessly), or indirectly (e.g., via a third component).

[0037] The methods according to various embodiments disclosed in this document may be provided as included in a computer program product. The computer program product may be traded as a commodity between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory, CD-ROM), or may be distributed online (e.g., downloaded or uploaded) through an application store or directly between two user devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

[0038] According to the embodiments disclosed in this document, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to the embodiments disclosed in this document, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to the embodiments disclosed in this document, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0039] FIG. 1 is a drawing illustrating elements constituting a battery diagnostic device according to some embodiments disclosed in this document.

[0040] Referring to FIG. 1, an electronic device (10) may include a battery (100) to be diagnosed, and a battery diagnostic device (200) may diagnose whether the battery (100) to be diagnosed is defective. The battery (100) to be diagnosed and the battery diagnostic device (200) may exchange data with each other via wired / wireless communication.

[0041] The electronic device (10) may refer to a device that uses power supplied by a battery (100) to be diagnosed. For example, the electronic device (10) may be an electric vehicle (EV) or a hybrid electric vehicle (HEV) that uses electric energy, and the battery (100) to be diagnosed may be a secondary battery mounted on the electric vehicle (EV) or the hybrid electric vehicle (HEV). Alternatively, the electronic device (10) may be a portable device such as a laptop computer, a smartphone, a tablet PC, etc., and may also be interpreted as various devices having a battery recharge function.

[0042] The battery (100) to be diagnosed may be an electrical energy supply device that supplies power to the electronic device (10). For example, the battery (100) to be diagnosed may be a rechargeable secondary battery that is discharged while supplying power to the electronic device (10) and charged by a battery charging device (not shown). According to some embodiments, the battery (100) to be diagnosed may include a battery cell, a battery module, a battery pack, and / or a battery rack.

[0043] The battery diagnostic device (200) may be configured to diagnose whether a cell, module, pack, and / or rack of a battery (100) to be diagnosed is defective. The battery diagnostic device (200) may include a sensor (210) and a processor (220). However, the present invention is not limited thereto, and some components may be omitted from the battery diagnostic device (200), and other general-purpose elements may be further included in the battery diagnostic device (200).

[0044] In some embodiments, the battery diagnostic device (200) may be configured as an integrated unit with a battery charging device (not shown). In some embodiments, at least a portion of the sensor (210) and the processor (220) may be mounted inside the electronic device (10) rather than the battery diagnostic device (200). That is, the functions performed by the battery diagnostic device (200) may be performed by a BMS (battery management system) inside the electric device (10), as well as in various devices such as a server, cloud, charger, or charger / discharger.

[0045] The battery to be diagnosed (100) and the battery diagnostic device (200) may be electrically connected to each other via a wired and / or wireless data communication network. For example, the wired network may be based on LAN (local area network) communication or power line communication, and the wireless network may be based on a short-range communication network such as Bluetooth, WiFi (wireless fidelity), IrDA (infrared data association), etc., and / or a long-range communication network such as a cellular network, 4G network, or 5G network. According to some embodiments, the communication method between the battery to be diagnosed (100) and the battery diagnostic device (200) may be a device-to-device communication method such as a bus, GPIO (general purpose input and output), SPI (serial peripheral interface), MIPI (mobile industry processor interface), etc.

[0046] The sensor (210) may be configured to measure battery test data from the battery to be diagnosed (100) during a charge / discharge test. According to an embodiment, the charge / discharge test may mean applying a specific type of voltage or current to check for defects in cells, modules, packs, etc. of the battery to be diagnosed (100). For example, the battery test data measured from the battery to be diagnosed (100) by the sensor (210) may include at least one of voltage data, current data, and temperature data of the battery to be diagnosed (100) measured during a time period of the charge / discharge test.

[0047] The sensor (210) may include at least one sensor for measuring battery-related data such as voltage, current, and / or temperature from the battery to be diagnosed (100).

[0048] The processor (220) may have a structure for executing commands that implement the operations of the battery diagnostic device (200). The processor (220) may be implemented as an array of multiple logic gates for processing various operations or as a general-purpose microprocessor, and may be composed of a single processor or multiple processors. For example, the processor (220) may be implemented in the form of at least one of a microprocessor, a CPU, a GPU, and an AP.

[0049] The processor (220) may be configured separately or integrally with a memory module (not shown) configured to store instructions, and may process various operations by executing instructions stored in the memory module. The memory module may store various data, instructions, mobile applications, computer programs, etc. For example, the memory module may be implemented as a non-volatile memory such as ROM, PROM, EPROM, EEPROM, flash memory, PRAM, MRAM, RRAM, FRAM, etc., or a volatile memory such as DRAM, SRAM, SDRAM, PRAM, RRAM, FeRAM, etc., and may be implemented in the form of an HDD, SSD, SD, Micro-SD, etc., or in the form of a combination thereof.

[0050] The processor (220) may be configured to divide battery test data into a plurality of charge / discharge state data sets based on a plurality of reference values ​​representing a plurality of charge / discharge states. In an embodiment, the plurality of reference values ​​may include voltage, current, or temperature values.

[0051] The battery (100) to be diagnosed may have one of a plurality of charge / discharge states. According to an embodiment, the type of charge / discharge state may be determined based on the value of the voltage or current applied to the battery (100) to be diagnosed. For example, if the battery test data is a voltage measured from the battery (100) to be diagnosed during a charge / discharge test section, the charge / discharge test section may be divided into a plurality of sections according to a specific criterion, and a specific charge / discharge state may be assigned to each of the plurality of sections. Here, the specific criterion for dividing the charge / discharge test section into a plurality of sections may be the same time interval, or other criteria may be utilized.

[0052] For example, when four charge / discharge states are utilized, four reference values ​​can be set, and the charge / discharge state of each section can be determined based on which of the four reference values ​​is closest to a representative value, such as an average voltage value of each section of a plurality of sections that divide the charge / discharge test sections. A set of sections having the same charge / discharge state can become a dataset of the corresponding charge / discharge state, and thus the battery test data can be divided into four charge / discharge state datasets. For more details, reference can be made to FIGS. 2 to 5, which will be described later.

[0053] The processor (220) may be configured to perform preliminary failure diagnosis on each of a plurality of charge / discharge state data sets based on a preliminary diagnosis model. Here, the preliminary diagnosis model may be any one of various predictive models for diagnosing battery failure, or a combination thereof. For example, the preliminary diagnosis model may include a support vector classifier (SVC) model, an extra-trees model, a cat boost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

[0054] The processor (220) can diagnose the state of the battery (100) to be diagnosed based on the result of the preliminary failure diagnosis. According to an embodiment, the processor (220) can be configured to determine the battery (100) to be diagnosed as a final failure if two or more data sets among a plurality of charge / discharge state data sets are determined as a preliminary failure as a result of the preliminary failure diagnosis. Here, the final failure may mean a failure according to an ensemble algorithm. For example, if four charge / discharge states are utilized, the battery test data can be divided into four charge / discharge state data sets, and four preliminary diagnosis results (failure diagnosis T or F) can be obtained for these. Here, the battery (100) to be diagnosed can be determined as a final failure only if two or more preliminary diagnosis results are determined as a failure (T). Meanwhile, the number of criteria for the final failure determination may be determined as one of two, three, and four depending on performance conditions such as the minimum selectivity and maximum over-inspection rate of the ensemble algorithm.

[0055] In the above aspect, the processor (220) may be understood to diagnose the state of the battery (100) to be diagnosed based on an ensemble algorithm. According to an embodiment, the ensemble algorithm may mean an algorithm that determines whether the battery (100) to be diagnosed is ultimately defective by using multiple diagnostic results for multiple charge / discharge state data sets of a preliminary diagnostic model.

[0056] In other words, the ensemble algorithm segments the fault diagnosis results into multiple charge / discharge states and utilizes their combinations. This allows for stable screening and over-detection rates even for typical charge / discharge patterns. To achieve this, preliminary fault diagnosis can be performed on each of the multiple charge / discharge state datasets using a preliminary diagnosis model.

[0057] According to some embodiments, the processor (220) may be further configured to calculate a diagnosis rate and an over-inspection rate that determine the battery (100) to be diagnosed as a final failure.

[0058] For example, a battery (100) to be diagnosed may include a plurality of battery cells, and it may be known in advance which of the plurality of battery cells are defective. The diagnosis rate may be calculated based on whether the ensemble algorithm correctly diagnoses whether the plurality of battery cells are defective, and the over-inspection rate may be calculated based on the ratio of non-defective cells to defective cells.

[0059] According to some embodiments, the plurality of charge / discharge states may include a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

[0060] For example, the above four charge and discharge states can indicate the current state of the battery (100) to be inspected, and can be distinguished based on the battery voltage. The charge state (Charge) can correspond to a high-intensity positive voltage, the post-charge rest state (Charge-rest) can correspond to a low-intensity positive voltage, the post-discharge rest state (Discharge-rest) can correspond to a high-intensity negative voltage, and the discharge state (Discharge) can correspond to a low-intensity negative voltage. According to some embodiments, the charge state (Charge) is distinguished into a rapid charge state (QC) and a constant-current charge state (CC), so that a total of five charge and discharge states can be utilized.

[0061] According to some embodiments, the processor (220) may be further configured to select, from among combinations of two or more states selected from a plurality of charge / discharge states, a combination that satisfies at least one of a condition in which a diagnostic rate is greater than or equal to a first threshold value and a condition in which an over-inspection rate is less than or equal to a second threshold value.

[0062] For example, when four charge / discharge states are utilized, there may be six cases for combinations of two charge / discharge states, four cases for combinations of three charge / discharge states, and one case for combinations of four charge / discharge states. Among these 11 combinations, a combination with high model performance can be selected based on the diagnosis rate and the over-detection rate. Specifically, when the preliminary diagnosis model is a support vector machine (SVM) model, and the combination of the charge-rest state and the discharge-rest state shows a high diagnosis rate and a low over-detection rate, the combination can be matched to the SVM model and stored in the battery diagnosis device (200).

[0063] In some embodiments, the preliminary diagnosis model may include at least one of a support vector classifier (SVC) model, an Extra-Trees model, a CatBoost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

[0064] The models exemplified above correspond to algorithms previously used for battery failure diagnosis and can be utilized as preliminary diagnostic models in ensemble algorithms. As described above, analysis can be conducted to determine which combinations of charge and discharge states correspond to each model. In addition to the exemplified models, various other algorithms and models applicable to battery failure diagnosis can also be utilized as preliminary diagnostic models.

[0065] FIG. 2 is a diagram illustrating battery test data according to some embodiments disclosed in this document.

[0066] Referring to FIG. 2, a graph (20) representing battery test data measured from a battery (100) to be diagnosed during a charge / discharge test may be illustrated. The battery test data of the graph (20) may be pre-processed data to which no processes for an ensemble algorithm have been applied. For example, in the graph (20), the horizontal axis (21) may represent time, and the vertical axis (22) may represent voltage.

[0067] The battery test data in graph (20) may be divided into multiple cycles. For example, each cycle may be one of the first through third cycles. However, the battery test data is not limited thereto, and may represent current or voltage and may consist of more or fewer cycles. Alternatively, the battery test data may have an atypical, general pattern rather than a combination of charge / discharge cycles.

[0068] FIG. 3 is a diagram illustrating a plurality of reference values ​​representing a plurality of charge / discharge states according to some embodiments disclosed in this document.

[0069] Referring to FIG. 3, a graph (30) may be illustrated that displays a plurality of reference values ​​representing a plurality of charge / discharge states. Each reference value of the graph (30) may mean a voltage level corresponding to each charge / discharge state.

[0070] For example, battery test data may be voltage data measured during a charge / discharge test, and four charge / discharge states may be utilized. A charge state reference value (31), a post-charge rest state reference value (32), a discharge state reference value (33), and a post-discharge rest state reference value (34) may be set for the four charge / discharge states. Each charge / discharge state may mean a specific range of voltage, and an average value, a median value, or other various types of representative values ​​representing the range may be set as the reference value.

[0071] FIG. 4 is a drawing illustrating a structure in which a charge state (Charge) is divided into a rapid charge state (QC) and a constant current charge state (CC) according to some embodiments disclosed in this document.

[0072] Referring to FIG. 4, a graph (40) may be illustrated showing a structure in which the state of charge (Charge) is divided into a rapid charge state (QC) and a constant current charge state (CC). In the graph (40), the state of charge reference value (31) may be subdivided into a rapid charge state reference value (41) and a constant current charge state reference value (42).

[0073] The state of charge (Charge) can be divided into a rapid charge state (QC) corresponding to a higher voltage range and a constant current charge state (CC) corresponding to a lower voltage range, and the reference value can also be subdivided accordingly. By subdividing the state of charge (Charge), which has the widest range among the four charge / discharge states consisting of the charge state (Charge-rest), the discharge state (Discharge), and the discharge-rest state (Discharge-rest), into two states, the ensemble algorithm can operate more precisely.

[0074] FIG. 5 is a diagram illustrating multiple charge / discharge state datasets according to some embodiments disclosed in this document.

[0075] Referring to FIG. 5, among five charge / discharge state data sets corresponding to five charge / discharge states, an equal current charge state data set (51) and a post-charge rest state data set (52) can be illustrated.

[0076] Among the five charge / discharge state datasets, three datasets not shown, namely the rapid charge state dataset, the discharge state dataset, and the post-discharge rest state dataset, can also be generated in the same manner. Each charge / discharge state dataset can be a subset of the battery test data. For example, combining the five charge / discharge state datasets can result in the battery test data of graph (20).

[0077] FIG. 6 is a diagram illustrating model performance of preliminary diagnostic models according to some embodiments disclosed in this document for each of a plurality of charge / discharge states.

[0078] Referring to FIG. 6, a table (60) may be illustrated that shows the diagnostic rates and over-diagnosis rates that the preliminary diagnostic models have for each of a plurality of charge / discharge states. The first column (61) of the table (60) may show the model name of the preliminary diagnostic model, the first row (62) of the table (60) may show the charge / discharge state, and the second row (63) of the table (60) may show the distinction between the diagnostic rate and the over-diagnosis rate.

[0079] The charge / discharge state of the first row (62) may mean three of the four charge / discharge states consisting of a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest). Alternatively, the charge / discharge state of the first row (62) may mean three of the five charge / discharge states consisting of a rapid charge state (QC), a constant current charge state (CC), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

[0080] For example, the preliminary diagnostic models in the first column (61) may refer to battery failure diagnostic models that have been used in the past. As in the value area of ​​the table (60), it can be confirmed that although some of the preliminary diagnostic models in the first column (61) show values ​​with high diagnostic rates, there is no model that has both a high diagnostic rate and a low over-inspection rate.

[0081] FIG. 7 is a diagram illustrating model performance of an ensemble algorithm according to some embodiments disclosed in this document.

[0082] Referring to FIG. 7, a table (70) may be illustrated to illustrate the model performance of an ensemble algorithm. For example, table (70) may represent an ensemble algorithm that determines a final failure if two or more of the three charge / discharge states in the first row (62) of table (60) are determined to be preliminary failures.

[0083] For example, among the three charge / discharge states in the first row (62), two or more states may be any one of the four combinations of (state 1, state 2), (state 1, state 3), (state 2, state 3), and (state 1, state 2, state 3) that exhibits the highest performance. As shown in Table (70), when the preliminary diagnosis model is “SVM2,” a diagnosis rate exceeding 90% and an over-detection rate below 10% can be achieved by the ensemble algorithm.

[0084] FIG. 8 is a diagram illustrating steps of a battery diagnosis method according to some embodiments disclosed in this document.

[0085] Referring to FIG. 8, the battery diagnosis method (800) may include steps (810) to (850). However, the present invention is not limited thereto, and some steps may be omitted or other general steps may be added, and the steps of the battery diagnosis method (800) may be executed in a different order than the illustrated order.

[0086] The battery diagnosis method (800) may be composed of steps that are processed in a time-series manner in the battery diagnosis device (200). Therefore, even if the details are omitted below, the details described above for the battery diagnosis device (200) may be equally applied to the battery diagnosis method (800).

[0087] Steps (810) to (850) of the battery diagnosis method (800) can be performed by the sensor (210) and processor (220) of the battery diagnosis device (200).

[0088] In step (810), the battery diagnostic device (200) can measure battery test data from the battery to be diagnosed through a sensor.

[0089] In step (820), the battery diagnostic device (200) can, through a processor, divide battery test data into a plurality of charge / discharge state data sets based on a plurality of reference values ​​associated with a plurality of charge / discharge states.

[0090] In step (830), the battery diagnostic device (200) can perform preliminary failure diagnosis for each of the plurality of charge / discharge state data sets based on the preliminary diagnosis model through the processor.

[0091] In step (840), the battery diagnosis device (200) can determine, through the processor, that the battery to be diagnosed is finally defective based on the result of the preliminary defect diagnosis.

[0092] Meanwhile, the battery diagnosis method (800) may be implemented in the form of a computer program stored on a computer-readable storage medium. That is, the computer program may include commands for implementing the battery diagnosis method (800), and the program commands may be stored on a computer-readable storage medium. The computer program may include a mobile application.

[0093] For example, a computer-readable storage medium may include magnetic media such as a hard disk, a floppy disk, and a magnetic tape, optical media such as a CD-ROM or a DVD, magneto-optical media such as a floptical disk, and hardware devices specially configured to store and execute computer program instructions, such as ROM, RAM, flash memory, and the like. The computer program instructions may include machine language code generated by a compiler and high-level language code that can be executed by a computer using an interpreter, etc.

[0094] The terms "include," "comprise," or "have" described above, unless otherwise specifically stated, imply that the corresponding component may be included, and therefore should be interpreted to include other components rather than to exclude other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments disclosed in this document belong, unless otherwise defined. Commonly used terms, such as terms defined in the dictionary, should be interpreted to be consistent with the contextual meaning of the relevant technology, and shall not be interpreted in an idealized or overly formal sense, unless explicitly defined in this document.

[0095] The above description is merely an example of the technical idea disclosed in this document, and those skilled in the art to which the embodiments disclosed in this document pertain may make various modifications and variations without departing from the essential characteristics of the embodiments disclosed in this document. Therefore, the embodiments disclosed in this document are not intended to limit the technical idea of ​​the embodiments disclosed in this document, but to explain it, and the scope of the technical idea disclosed in this document is not limited by these embodiments. The scope of protection of the technical idea disclosed in this document should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of rights of this document.

[0096] [Explanation of symbols]

[0097] 10: Electronic Device 100: Battery to be Diagnosed

[0098] 200: Battery diagnostic device 210: Sensor

[0099] 220: Processor

Claims

1. A sensor configured to measure battery test data from a battery to be diagnosed; and A battery diagnosis device comprising a processor configured to divide the battery test data into a plurality of charge / discharge state data sets based on a plurality of reference values ​​associated with a plurality of charge / discharge states, perform a preliminary failure diagnosis on each of the plurality of charge / discharge state data sets based on a preliminary diagnosis model, and determine the battery to be diagnosed as finally defective based on the result of the preliminary failure diagnosis.

2. In paragraph 1, When the above processor determines that the final defect is present, A battery diagnosis device configured to determine the battery to be diagnosed as a final defect if two or more of the plurality of charge / discharge state data sets are determined to be a preliminary defect.

3. In paragraph 1, A battery diagnostic device, wherein the processor is further configured to compare the result of determining the battery to be diagnosed as being defective with the actual defect of the battery to be diagnosed to calculate a diagnosis rate and an over-inspection rate.

4. In paragraph 3, A battery diagnostic device, wherein the plurality of charge and discharge states include a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

5. In paragraph 4, A battery diagnostic device, wherein the processor is further configured to select a combination that satisfies at least one of a condition in which the diagnostic rate is equal to or greater than a first threshold value and a condition in which the over-inspection rate is equal to or less than a second threshold value, among combinations composed of two or more states selected from among the plurality of charge / discharge states.

6. In paragraph 1, A battery diagnostic device, wherein the battery test data includes at least one of voltage data, current data, and temperature data of the battery to be diagnosed measured during a time period of a charge / discharge test.

7. In paragraph 1, A battery diagnostic device, wherein the preliminary diagnostic model comprises at least one of a support vector classifier (SVC) model, an extra-trees model, a CatBoost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

8. A step of measuring battery test data from a battery to be diagnosed through a sensor; A step of dividing the battery test data into a plurality of charge / discharge state data sets based on a plurality of reference values ​​associated with a plurality of charge / discharge states through a processor; A step of performing preliminary failure diagnosis for each of the plurality of charge / discharge state data sets based on a preliminary diagnosis model through the above processor; and A battery diagnosis method, comprising a step of determining the battery to be diagnosed as finally defective based on the result of the preliminary defective diagnosis through the processor.

9. In paragraph 8, The steps for determining the final defect above are: A battery diagnosis method, comprising a step of determining the battery to be diagnosed as a final defect when two or more of the plurality of charge / discharge status data sets are determined to be a preliminary defect.

10. In paragraph 8, A battery diagnosis method further comprising a step of comparing the result of determining the final defect in the above-mentioned battery to the actual defect of the above-mentioned battery to calculate a diagnosis rate and an over-inspection rate.

11. In paragraph 10, A battery diagnosis method, wherein the plurality of charge and discharge states include a charge state (Charge), a charge-rest state (Charge-rest), a discharge state (Discharge), and a discharge-rest state (Discharge-rest).

12. In paragraph 11, A battery diagnosis method further comprising a step of selecting a combination that satisfies at least one of a condition in which the diagnosis rate is equal to or greater than a first threshold value and a condition in which the over-inspection rate is equal to or less than a second threshold value, among combinations composed of two or more states selected from the plurality of charge / discharge states.

13. In paragraph 8, A battery diagnosis method, wherein the battery test data includes at least one of voltage data, current data, and temperature data of the battery to be diagnosed measured during a time period of a charge / discharge test.

14. In paragraph 8, A battery diagnosis method, wherein the preliminary diagnosis model comprises at least one of a support vector classifier (SVC) model, an extra-trees model, a CatBoost model, a support vector machine (SVM) model, and a long short-term memory (LSTM) model.

Images