Battery diagnosis device and method

Abstract

A battery diagnosis device according to an embodiment of the present invention may comprise: at least one processor; and memory that stores at least one command that is executed through the at least one processor. Here, the at least one command may comprise: a command for acquiring state data including a state value of each of factors for a plurality of batteries; a command for deriving regression lines indicating relationships between the factors by using the state data; a command for calculating coordinates corresponding to the state values of the factors and distances between the regression lines; and a command for inputting diagnostic data including the distances into a pre-trained diagnostic model, and determining an abnormal battery among the batteries by using a diagnostic result output by the diagnostic model.

Description

Battery diagnostic device and method

[0001] This application claims the benefit of the filing date of Korean Patent Application No. 10-2024-0132588 filed with the Korean Intellectual Property Office on September 30, 2024, and all contents disclosed in the document of said Korean patent application are incorporated into this specification.

[0002] The present invention relates to a battery diagnostic device and method, and more specifically, to a battery diagnostic device and method that diagnoses whether there is an abnormality in a battery using an artificial intelligence-based diagnostic model.

[0003] Secondary batteries are batteries that can be reused through charging even after discharge, and can be used as an energy source for small devices such as mobile phones, tablet PCs, and vacuum cleaners, and are also used as an energy source for medium and large devices such as automobiles and Energy Storage Systems (ESS) for smart grids.

[0004] Secondary batteries are applied to systems in the form of assemblies, such as battery modules in which multiple battery cells are connected in series and parallel, or battery packs in which battery modules are connected in series and parallel, depending on the system's requirements. For medium to large-sized devices, such as electric vehicles, a high-capacity battery system in which multiple battery packs are connected in parallel may be applied to satisfy the device's required capacity.

[0005] Battery cells are manufactured through assembly and activation processes. Since battery cells are assembled in a discharged state, they must undergo an activation process after assembly to activate the positive electrode active material and form a Solid Electrolyte Interface (SEI) on the negative electrode, thereby enabling the cell to function as a battery. This activation process is called the formation process. The activation process can be carried out by a charge / discharge device that repeatedly charges and discharges the battery cell according to a preset charge / discharge profile.

[0006] Generally, to identify defective batteries during the manufacturing process, status data such as voltage and current values ​​of battery cells are collected during the activation process, and defective batteries can be identified using the collected status data. For example, low-voltage defective batteries can be detected based on the voltage drop measured during the activation process.

[0007] However, according to this general diagnostic method, a normal battery may be identified as a defective battery (over-inspected), or a defective battery may be judged as a normal battery (under-inspected).

[0008] A related prior art is KR 10-2022-0039452 A.

[0009] The objective of the present invention, which aims to solve the aforementioned problems, is to provide a battery diagnostic device capable of diagnosing whether a battery is abnormal using an artificial intelligence-based diagnostic model.

[0010] Another objective of the present invention to solve the above-mentioned problems is to provide a battery diagnosis method using such a battery diagnosis device.

[0011] Another objective of the present invention to solve the above-mentioned problems is to provide a battery system including such a battery diagnostic device.

[0012] A battery diagnostic device according to one embodiment of the present invention for achieving the above objective may include at least one processor; and a memory that stores at least one instruction executed through the at least one processor.

[0013] Herein, the at least one command may include: a command to acquire state data including state values ​​of each of the factors for a plurality of batteries; a command to derive a regression line representing the relationship between the factors using the state data; a command to calculate the distance between the coordinates corresponding to the state values ​​of the factors and the regression line; and a command to input diagnostic data including the distance into a pre-trained diagnostic model and to determine an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

[0014] The above factors may include one or more of the battery's voltage, current, temperature, and capacity.

[0015] The command for deriving the regression line may include a command for deriving a regression line representing the relationship between the two factors by performing a predefined linear regression analysis using state values ​​for each of the two factors.

[0016] The command for calculating the distance may include a command for calculating the vertical distance between the two-dimensional coordinates corresponding to the state values ​​of each of the two factors and the regression line.

[0017] The command for deriving the regression line may include, when there are three or more factors, a command for deriving multiple combinations consisting of two factors; and a command for deriving a regression line for each of the combinations.

[0018] The above diagnostic model can be pre-trained to output diagnostic result data indicating whether there is a low voltage defect of a specific battery when receiving diagnostic data for that battery.

[0019] The above diagnostic data may further include one or more of the average value of the distances for the plurality of batteries and the standard deviation value of the distances.

[0020]

[0021] A battery diagnostic method according to an embodiment of the present invention for achieving the above other objectives may include: a battery diagnostic method using a battery diagnostic device, comprising the steps of: acquiring state data including state values ​​of each of the factors for a plurality of batteries; deriving a regression line representing the relationship between the factors using the state data; calculating the distance between the coordinates corresponding to the state values ​​of the factors and the regression line; and inputting the diagnostic data including the distance into a pre-trained diagnostic model and determining an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

[0022] The above factors may include one or more of the battery's voltage, current, temperature, and capacity.

[0023] The step of deriving the regression line may include the step of performing a predefined linear regression analysis using state values ​​for each of the two factors to derive a regression line representing the relationship between the two factors.

[0024] The step of calculating the distance may include the step of calculating the vertical distance between the two-dimensional coordinates corresponding to the state values ​​of each of the two factors and the regression line.

[0025] The step of deriving the regression line may include, when there are three or more factors, a step of deriving a plurality of combinations consisting of two factors; and a step of deriving a regression line for each of the combinations.

[0026] The above diagnostic model can be pre-trained to output diagnostic result data indicating whether there is a low voltage defect of a specific battery when receiving diagnostic data for that battery.

[0027] The above diagnostic data may further include one or more of the average value of the distances for the plurality of batteries and the standard deviation value of the distances.

[0028]

[0029] A battery system according to one embodiment of the present invention for achieving the above-mentioned other purpose may include: a data collection device for collecting state data including state values ​​of each of the factors for a plurality of batteries; and a battery diagnostic device for obtaining the state data from the data collection device and diagnosing whether the batteries are abnormal using the state data.

[0030] Here, the battery diagnostic device can derive a regression line representing the relationship between the factors using the state data, calculate the distance between the coordinates corresponding to the state values ​​of the factors and the regression line, and determine an abnormal battery among the batteries using the calculated distance and a pre-trained diagnostic model.

[0031] According to the embodiment of the present invention as described above, diagnostic accuracy can be improved by detecting abnormal batteries using an artificial intelligence-based diagnostic model.

[0032] Figure 1 is a reference diagram for explaining a general battery diagnostic method.

[0033] FIG. 2 is a block diagram showing a battery diagnostic system according to an embodiment of the present invention.

[0034] FIG. 3 is a block diagram showing a battery diagnostic device and a diagnostic model generation device according to an embodiment of the present invention.

[0035] FIGS. 4 to 7 are reference diagrams for explaining a diagnostic model according to an embodiment of the present invention.

[0036] FIG. 8 is a flowchart of the operation sequence of a battery diagnostic method according to an embodiment of the present invention.

[0037] FIG. 9 is a block diagram of a battery diagnostic device according to an embodiment of the present invention.

[0038] 10: Battery

[0039] 100: Data acquisition device

[0040] 200, 900: Battery diagnostic device

[0041] 300: Diagnostic model generation device

[0042] The present invention is susceptible to various modifications and may have various embodiments; specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. Similar reference numerals have been used for similar components in the description of each drawing.

[0043] Terms such as first, second, A, B, etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0044] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0045] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0046] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0047]

[0048] Some terms used in this specification are defined as follows.

[0049] SOC (State of Charge) is the battery's current charged state expressed as a percentage, and SOH (State of Health) is the battery's current degradation state expressed as a percentage.

[0050] A battery cell is the smallest unit that performs the role of storing electricity, and a battery module refers to an assembly of multiple battery cells that are electrically connected.

[0051] A battery rack refers to a single-structure system in which module units set by the battery manufacturer are electrically connected and can be monitored and controlled through a BMS, and may be configured to include multiple battery modules and one BPU or protection device. Here, depending on the device or system in which the battery is used, the battery modules may also be referred to as battery packs.

[0052] A battery bank may refer to a group of large-scale battery rack systems configured by connecting multiple battery racks in parallel. Monitoring and control of a rack BMS (RBMS) at the battery rack level can be performed through a battery bank-level BMS.

[0053] A battery assembly refers to a collection comprising a plurality of electrically connected battery cells that is applied to a specific system or device and functions as a power source. Here, a battery assembly may refer to a battery module, a battery pack, a battery rack, or a battery bank, but the scope of the present invention is not limited to these entities.

[0054]

[0055] Figure 1 is a reference diagram for explaining a general battery diagnostic method.

[0056] Generally, to identify defective batteries during the manufacturing process, state data such as voltage and current values ​​of battery cells are collected during the activation process, and defective batteries can be identified using the collected state data.

[0057] For example, a battery diagnostic device can collect the open circuit voltage (OCV) of each battery cell measured during the activation process. Subsequently, as shown in FIG. 1, the battery diagnostic device can identify battery cells in which the voltage drop exceeds a preset threshold value and determine the identified battery cells as low-voltage defective battery cells.

[0058] This general diagnostic method is based on statistics of specific status values ​​and exhibits limitations in diagnostic accuracy as it does not consider various factors. Consequently, according to this method, there may be cases where a normal battery is identified as defective (over-inspection) or a defective battery is judged as normal (under-inspection).

[0059] The present invention is a technology devised to solve these problems, and various embodiments of the present invention will be described in detail below with reference to the attached drawings.

[0060]

[0061] FIG. 2 is a block diagram showing a battery diagnostic system according to an embodiment of the present invention.

[0062] Referring to FIG. 2, the battery diagnostic system may include a plurality of batteries (10) (BAT #1 to BAT #N), a data collection device (100), a battery diagnostic device (200), and a diagnostic model generation device (300).

[0063] In the present invention, the battery (10) refers to a battery cell, but is not limited thereto. That is, the battery diagnostic device (200) according to the present invention can perform diagnostics on a battery module, battery pack, battery rack, or battery bank unit, in addition to a battery cell.

[0064] Each of the batteries (10) can be electrically connected to a charging / discharging device or a charging / discharging circuit. Here, the batteries (10) can be charged according to a predefined charging profile or charging map.

[0065] The data collection device (100) can collect status data for each of the batteries.

[0066] The state data may include the state values ​​of each of a plurality of factors. Here, the factors may include one or more of the voltage, current, temperature, and capacity of the battery. For example, the data collection device (100) may collect state data including one or more of the OCV value, current value, temperature value, and capacity value of each of the battery cells measured during the activation process. As another example, the data collection device (100) may collect state data including one or more of the OCV value, current value, temperature value, and capacity value of each of the battery modules measured during the operation of the battery system.

[0067] The data collection device (100) can provide the collected status data to the battery diagnostic device (200).

[0068] The battery diagnostic device (200) can diagnose whether each of the batteries (10) is abnormal by using battery status data provided from the data collection device (100). Here, the battery diagnostic device (200) can detect a battery among the plurality of batteries (10) in which a predefined abnormal state has occurred. For example, the battery diagnostic device (200) can detect a low-voltage defective battery among the plurality of batteries (10).

[0069] The battery diagnostic device (200) may be included in the activation process equipment or in a battery management system (BMS) located inside the battery system. That is, the battery diagnostic system according to an embodiment of the present invention may be applied to the activation process to detect defective battery cells during the battery manufacturing process, or applied to the battery system to detect abnormal battery cells or abnormal battery assemblies during the operation of the battery system.

[0070] The battery diagnostic device (200) can generate diagnostic data according to a predefined data processing process for state data and determine an abnormal battery among the batteries using the diagnostic data and a pre-trained diagnostic model.

[0071] A diagnostic model can be generated by a diagnostic model generating device (300). Here, the diagnostic model generating device (300) can train a predefined diagnostic model using pre-collected training data and provide the trained diagnostic model to a battery diagnostic device (200). Here, the diagnostic model can be pre-trained to output diagnostic result data indicating whether there is a low voltage defect of the battery when it receives diagnostic data for a specific battery.

[0072] The battery diagnostic device (200) can input diagnostic data into a diagnostic model and determine an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

[0073]

[0074] FIG. 3 is a block diagram showing a battery diagnostic device and a diagnostic model generation device according to an embodiment of the present invention.

[0075] The battery diagnostic device (200) can determine an abnormal battery among the batteries using a pre-trained diagnostic model (210). Here, the battery diagnostic device (200) can receive a diagnostic model (310) that has been learned from the diagnostic model generation device (200).

[0076] The battery diagnostic device (200) can obtain state data including state values ​​of each factor for the batteries from the data collection device (100) of FIG. 2, and process the state data according to a predefined data processing process to generate input data (diagnostic data) applied to the diagnostic model (210). Subsequently, the battery diagnostic device (200) can input the diagnostic data into the diagnostic model (210) and determine an abnormal battery among the batteries using the diagnostic result (output data) output from the diagnostic model (210).

[0077] The diagnostic model generation device (300) can train a machine learning-based diagnostic model (310) using training data stored in a storage device (320). Here, the diagnostic model (310) can be trained to output diagnostic result data indicating whether there is a low voltage defect of the battery when it receives diagnostic data for a specific battery.

[0078] When the learning of the diagnostic model (310) is completed, the diagnostic model generating device (300) can provide the learned diagnostic model (310) to the battery diagnostic device (200).

[0079]

[0080] FIGS. 4 to 7 are reference diagrams for explaining a diagnostic model according to an embodiment of the present invention.

[0081] The diagnostic model generating device (300) can collect raw data to train the diagnostic model (310). Here, the raw data may include state data containing state values ​​of each factor for a plurality of batteries and diagnostic result data indicating whether there is an abnormality in the battery, and the diagnostic model generating device (300) can obtain raw data from the data collection device (100).

[0082] The factors may include one or more of the battery voltage, current, temperature, and capacity. For example, the state data may include a first OCV value (state value of the first factor) of each battery cell measured after the completion of a first charging cycle during the activation process, and a second OCV value (state value of the second factor) of each battery cell measured after the completion of a second charging cycle during the activation process. As another example, the state data may include the OCV value (state value of the first factor), current value (state value of the second factor), temperature value (state value of the third factor), and capacity value (state value of the fourth factor) of each battery cell measured during the activation process. As yet another example, the state data may include the OCV value (state value of the first factor), current value (state value of the second factor), temperature value (state value of the third factor), and capacity value (state value of the fourth factor) of each battery module measured during the operation of the battery system.

[0083] The diagnostic model generation device (300) can generate training data for learning the diagnostic model (310) by processing raw data according to a predefined data processing process. Here, the training data may include coordinates corresponding to the state values ​​of factors (hereinafter, coordinates of the state values), distances between regression lines representing the relationship between factors, and diagnostic result data representing whether the battery is abnormal (e.g., whether there is a low voltage defect).

[0084] Specifically, the diagnostic model generating device (300) can derive a regression line representing the relationship between multiple factors using state data. Here, the diagnostic model generating device (300) can perform a predefined linear regression analysis using state values ​​for each of the two factors and derive a regression line representing the relationship between the two factors.

[0085] For example, if the raw data includes a first OCV value (state value of the first factor) measured after the completion of the first charging cycle during the activation process for each of the plurality of battery cells, and a second OCV value (state value of the second factor) measured after the completion of the second charging cycle during the activation process, the diagnostic model generating device (300) can define the first factor as the X-axis and the second factor as the Y-axis, as shown in FIG. 4, and convert the state value of the first factor and the state value of the second factor into two-dimensional coordinates for each of the battery cells. Subsequently, the diagnostic model generating device (300) can derive a regression line (Lr) that estimates the relationship between the first factor and the second factor by performing a predefined linear regression analysis using the coordinates of the state values ​​for the battery cells, as shown in FIG. 5.

[0086] Subsequently, the diagnostic model generating device (300) can calculate the distance between the coordinates of the state value and the regression line for each of the battery cells. Here, the diagnostic model generating device (300) can calculate the vertical distance (D) between the two-dimensional coordinates corresponding to the state value of each of the two factors and the regression line (Lr).

[0087] For example, the diagnostic model generating device (300) can calculate the vertical distance (or shortest distance) (D) between the coordinates of the state value and the regression line (Lr) for a specific battery cell (Bat #23), as shown in FIG. 6. Here, the diagnostic model generating device (300) can calculate the vertical distance (D) for each of all battery cells.

[0088] The diagnostic model generation device (300) can train a diagnostic model (310) using training data including the distance (D) between the coordinates of the state values ​​and the regression line for a plurality of batteries, and diagnostic result data indicating whether the battery is abnormal.

[0089] In an embodiment, when state values ​​for three or more factors are included in the raw data, the diagnostic model generating device (300) may derive a plurality of combinations consisting of two factors and derive a regression line for each of the combinations. Subsequently, the diagnostic model generating device (300) may calculate the distance (D1 to DN) between the coordinates of the state value and the regression line for each of the regression lines. Here, the distance (D1 to DN) between the coordinates of the state value and the regression line calculated for each of the regression lines may be included in the training data.

[0090] For example, if the raw data includes an OCV value (state value of the first factor) measured after the completion of the activation process for each of the plurality of battery cells, a current value (state value of the second factor) measured at a specific point in time during the activation process, and a capacity value (state value of the third factor) measured after the completion of the activation process, the diagnostic model generating device (300) can derive from the three factors a plurality of combinations consisting of two factors, such as [first factor, second factor], [first factor, third factor], and [second factor, third factor].

[0091] In another example, if the raw data includes, for each of the plurality of battery cells, an OCV value (state value of the first factor) measured at a first time point after the completion of the activation process, an OCV value (state value of the second factor) measured at a second time point (a time point after t period elapsed from the first time point) after the completion of the activation process, and an OCV value (state value of the third factor) measured at a third time point (a time point after t period elapsed from the second time point) after the completion of the activation process, the diagnostic model generating device (300) can derive from the three factors a plurality of combinations consisting of two factors, such as [first factor, second factor], [first factor, third factor], and [second factor, third factor].

[0092] Subsequently, the diagnostic model generating device (300) can derive a first regression line (Lr1) representing the relationship between a first factor and a second factor as illustrated in FIG. 7(A), and calculate a first distance (D1) between the state value coordinates of the first factor and the second factor and the first regression line (Lr1) for all battery cells. Additionally, the diagnostic model generating device (300) can derive a second regression line (Lr2) representing the relationship between a first factor and a third factor as illustrated in FIG. 7(B), and calculate a second distance (D2) between the state value coordinates of the first factor and the third factor and the second regression line (Lr2) for all battery cells. Additionally, the diagnostic model generating device (300) can derive a third regression line (Lr3) representing the relationship between the second factor and the third factor as illustrated in FIG. 7(C), and calculate a third distance (D3) between the state value coordinates of the second factor and the third factor and the third regression line (Lr3) for all battery cells.

[0093] Subsequently, the diagnostic model generating device (300) can train the diagnostic model (310) using training data including a first distance (D1), a second distance (D2), a third distance (D3) for a plurality of batteries, and diagnostic result data indicating whether there is an abnormality in the battery.

[0094] In an embodiment, the training data may further include one or more of the average value (Av) of the distances (D) between the coordinates of the state value and the regression line, and the standard deviation value (S) of the distances (D).

[0095] For example, the diagnostic model generating device (300) can calculate a distance average value (Av) and a standard deviation value (S) based on the distance (D) from the regression line for each of the plurality of batteries, and train the diagnostic model (310) by reflecting the calculated distance average value (Av) and standard deviation value (S) in the training data.

[0096] In another example, the diagnostic model generating device (300) can calculate a first average distance value (Av1) and a first standard deviation value (S1) based on the state value coordinates of a first factor and a second factor for each of the plurality of batteries and a first distance (D1) between the first regression line. Additionally, the diagnostic model generating device (300) can calculate a second average distance value (Av2) and a second standard deviation value (S2) based on the state value coordinates of a first factor and a third factor for each of the plurality of batteries and a second distance (D2) between the second regression line. Additionally, the diagnostic model generating device (300) can calculate a third average distance value (Av3) and a third standard deviation value (S3) based on the state value coordinates of a second factor and a third factor for each of the plurality of batteries and a third distance (D3) between the third regression line. Afterwards, the diagnostic model generating device (300) can train the diagnostic model (310) by reflecting the first to third distance average values ​​(Av1 to Av3) and the first to third standard deviation values ​​(S1 to S3) in the training data.

[0097] When the learning of the diagnostic model (310) is completed, the diagnostic model generating device (300) can provide the learned diagnostic model (310) to the battery diagnostic device (200).

[0098]

[0099] FIG. 8 is a flowchart of the operation sequence of a battery diagnostic method according to an embodiment of the present invention.

[0100] In the following description, a battery diagnostic method targeting a single battery is described as an example; however, the battery diagnostic method according to an embodiment of the present invention can be performed on each of a plurality of batteries, and accordingly, an abnormal battery among the plurality of batteries can be detected.

[0101] The battery diagnostic device can acquire status data including the status values ​​of each of the factors for a plurality of batteries (S810). Here, the battery diagnostic device can acquire status data for each of the batteries from a data collection device.

[0102] Factors may include one or more of the battery's voltage, current, temperature, and capacity.

[0103] A battery diagnostic device can collect state data corresponding to the training data of a diagnostic model. For example, if the diagnostic model is pre-trained based on a first OCV value (state value of the first factor) of each battery cell measured after the completion of a first charging cycle during the activation process and a second OCV value (state value of the second factor) of each battery cell measured after the completion of a second charging cycle during the activation process, the battery diagnostic device can collect state data including the state value of the first factor and the state value of the second factor. As another example, if the diagnostic model is pre-trained based on the OCV value (state value of the first factor), current value (state value of the second factor), and capacity value (state value of the third factor) of each battery cell measured during the activation process, the battery diagnostic device can collect state data including the state value of the first factor, the state value of the second factor, and the state value of the third factor.

[0104] The battery diagnostic device can derive a regression line representing the relationship between factors using the status data obtained in S810 (S820). Here, the battery diagnostic device can perform a predefined linear regression analysis using the status values ​​for each of the two factors and derive a regression line representing the relationship between the two factors.

[0105] For example, if the state data includes a first OCV value (state value of the first factor) measured after the completion of the first charging cycle during the activation process for each of the plurality of battery cells, and a second OCV value (state value of the second factor) measured after the completion of the second charging cycle during the activation process, the battery diagnostic device can define the first factor as the X-axis and the second factor as the Y-axis, as shown in FIG. 4, and convert the state value of the first factor and the state value of the second factor into two-dimensional coordinates for each of the battery cells. Subsequently, the battery diagnostic device can derive a regression line (Lr) that estimates the relationship between the first factor and the second factor by performing a predefined linear regression analysis using the coordinates of the state values ​​for the battery cells, as shown in FIG. 5.

[0106] A battery diagnostic device can calculate the distance between the coordinates of a state value and a regression line for a specific battery (S830). Here, the battery diagnostic device can calculate the vertical distance (D) between the two-dimensional coordinates corresponding to the state value of each of the two factors and the regression line (Lr).

[0107] For example, the battery diagnostic device can calculate the vertical distance (or shortest distance) (D) between the coordinates of the state value and the regression line (Lr) for a specific battery cell (Bat #23), as shown in FIG. 6.

[0108] The battery diagnostic device can determine whether the battery is abnormal by using the distance (D) between the coordinates of the state value calculated in S830 and the regression line, and a pre-learned diagnostic model (S840).

[0109] More specifically, the battery diagnostic device may input diagnostic data including the distance (D) calculated in S830 into a pre-stored diagnostic model. Here, the diagnostic model may be stored in the storage device of the battery diagnostic device in a pre-trained state to output diagnostic result data (e.g., OK or NG) indicating whether the battery is abnormal (e.g., whether there is a low voltage defect) when it receives diagnostic data for a specific battery. Accordingly, the battery diagnostic device can determine whether the battery is abnormal by checking the diagnostic result data output by the diagnostic model.

[0110]

[0111] In an embodiment, when state values ​​for three or more factors are included in the state data, the battery diagnostic device may derive multiple combinations consisting of two factors and derive a regression line for each of the combinations. Subsequently, for each of the regression lines, the battery diagnostic device may calculate the distance (D1 ~ DN) between the coordinates of the state value and the regression line.

[0112] For example, if the state data includes an OCV value (state value of the first factor) measured after the completion of the activation process for each of the plurality of battery cells, a current value (state value of the second factor) measured at a specific point in time during the activation process, and a capacity value (state value of the third factor) measured after the completion of the activation process, the battery diagnostic device can derive from the three factors a plurality of combinations consisting of two factors, such as [first factor, second factor], [first factor, third factor], and [second factor, third factor].

[0113] In another example, if the state data includes, for each of the plurality of battery cells, an OCV value (state value of the first factor) measured at a first time point after the completion of the activation process, an OCV value (state value of the second factor) measured at a second time point (a time point after t period elapsed from the first time point) after the completion of the activation process, and an OCV value (state value of the third factor) measured at a third time point (a time point after t period elapsed from the second time point) after the completion of the activation process, the battery diagnostic device can derive from the three factors a plurality of combinations consisting of two factors, such as [first factor, second factor], [first factor, third factor], and [second factor, third factor].

[0114] Subsequently, the battery diagnostic device can derive a first regression line (Lr1) representing the relationship between a first factor and a second factor, as illustrated in FIG. 7(A), and calculate a first distance (D1) between the state value coordinates of the first factor and the second factor and the first regression line (Lr1). Additionally, the battery diagnostic device can derive a second regression line (Lr2) representing the relationship between a first factor and a third factor, as illustrated in FIG. 7(B), and calculate a second distance (D2) between the state value coordinates of the first factor and the third factor and the second regression line (Lr2). Additionally, the battery diagnostic device can derive a third regression line (Lr3) representing the relationship between a second factor and a third factor, as illustrated in FIG. 7(C), and calculate a third distance (D3) between the state value coordinates of the second factor and the third factor and the third regression line (Lr3).

[0115] Subsequently, the battery diagnostic device inputs diagnostic data including first to third distances (D1 to D3) into a pre-stored diagnostic model and checks the diagnostic result data output by the diagnostic model to determine whether the battery is abnormal.

[0116]

[0117] In another embodiment, the diagnostic data may further include one or more of the average value (Av) of the distances (D) between the coordinates of the status value and the regression line, and the standard deviation value (S) of the distances (D).

[0118] For example, the battery diagnostic device may calculate an average distance value (Av) and a standard deviation value (S) based on the distance (D) from the regression line for each of the multiple batteries, and include the calculated average distance value (Av) and standard deviation value (S) in the diagnostic data. Subsequently, the battery diagnostic device may input the diagnostic data, which includes the distance (D) between the coordinates of the status value for a specific battery and the regression line, as well as the average distance value (Av) and standard deviation value (S), into a diagnostic model, and determine whether the battery is abnormal by checking the diagnostic result data output by the diagnostic model.

[0119]

[0120] FIG. 9 is a block diagram of a battery diagnostic device according to an embodiment of the present invention.

[0121] A battery diagnostic device (900) according to an embodiment of the present invention may include at least one processor (910), a memory (920) that stores at least one command executed through the processor, and a transceiver (930) that is connected to a network to perform communication.

[0122] The above at least one command may include: a command to acquire state data including state values ​​of each of the factors for a plurality of batteries; a command to derive a regression line representing the relationship between the factors using the state data; a command to calculate the distance between the coordinates corresponding to the state values ​​of the factors and the regression line; and a command to input diagnostic data including the distance into a pre-trained diagnostic model and to determine an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

[0123] The above factors may include one or more of the battery's voltage, current, temperature, and capacity.

[0124] The command for deriving the regression line may include a command for deriving a regression line representing the relationship between the two factors by performing a predefined linear regression analysis using state values ​​for each of the two factors.

[0125] The command for calculating the distance may include a command for calculating the vertical distance between the two-dimensional coordinates corresponding to the state values ​​of each of the two factors and the regression line.

[0126] The command for deriving the regression line may include, when there are three or more factors, a command for deriving multiple combinations consisting of two factors; and a command for deriving a regression line for each of the combinations.

[0127] The above diagnostic model can be pre-trained to output diagnostic result data indicating whether there is a low voltage defect of a specific battery when receiving diagnostic data for that battery.

[0128] The above diagnostic data may further include one or more of the average value of the distances for the plurality of batteries and the standard deviation value of the distances.

[0129] The battery diagnostic device (900) may also further include an input interface device (940), an output interface device (950), a storage device (960), etc. Each component included in the battery diagnostic device (900) can be connected by a bus (970) to communicate with each other.

[0130] Here, the processor (910) may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. Additionally, the memory may be composed of at least one of a volatile / transitory storage medium and a non-volatile / non-transitory storage medium. For example, the memory may be composed of at least one of a read-only memory (ROM) and a random access memory (RAM), and may include an EEPROM (Electrically Erasable Programmable Read-only Memory).

[0131] The operation of the method according to an embodiment of the present invention can be implemented as a computer-readable program or code on a computer-readable recording medium. The computer-readable recording medium may include any type of recording device in which data that can be read by a computer system is stored. The computer-readable recording medium may also be distributed across networked computer systems, so that the computer-readable program or code can be stored and executed in a distributed manner.

[0132] The operation of the method according to an embodiment of the present invention can be implemented in various forms related to a program, such as a computer program or code itself or a computer program product.

[0133] Additionally, computer-readable recording media may include one or more of volatile / transitory recording media and non-volatile / non-transitory recording media.

[0134] Computer-readable recording media may include hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory, and may include, for example, various types of servers located on a network. Program instructions may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0135] Some aspects of the invention have been described in the context of a device, but may also be described according to a corresponding method, wherein a block or device corresponds to a method step or a feature of a method step. Similarly, aspects described in the context of a method may also be described according to a corresponding block or item or a feature of a corresponding device. Some or all of the method steps may be performed by (or using) a hardware device, such as, for example, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, one or more of the most important method steps may be performed by such a device.

[0136] Although the present invention has been described with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as described in the following claims.

Claims

1. At least one processor; and It includes a memory that stores at least one instruction executed through the above-mentioned at least one processor, and The above at least one command is, A command to acquire state data including the state value of each of the factors for a plurality of batteries; A command to derive a regression line representing the relationship between the factors using the above state data; A command to calculate the distance between the coordinates corresponding to the state values ​​of the above factors and the regression line; and A battery diagnostic device comprising a command to input diagnostic data including the above distance into a pre-trained diagnostic model, and to determine an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

2. In Claim 1, The above factor is, A battery diagnostic device comprising one or more of the voltage, current, temperature, and capacity of a battery.

3. In Claim 1, The command to derive the above regression line is, A battery diagnostic device comprising a command to perform a predefined linear regression analysis using state values ​​for each of two factors to derive a regression line representing the relationship between the two factors.

4. In Claim 3, The command to calculate the above distance is, A battery diagnostic device comprising a command to calculate the vertical distance between the two-dimensional coordinates corresponding to the state values ​​of each of the two factors and the regression line.

5. In Claim 3, The command to derive the above regression line is, A command to derive multiple combinations consisting of two factors when the above factors are three or more; and A battery diagnostic device comprising a command to derive a regression line for each of the above combinations.

6. In Claim 1, The above diagnostic model is, A battery diagnostic device that is pre-trained to output diagnostic result data indicating whether there is a low voltage defect of a specific battery when receiving diagnostic data for that specific battery.

7. In Claim 1, The above diagnostic data is, A battery diagnostic device comprising one or more of the average value of the distances for the plurality of batteries and the standard deviation value of the distances.

8. A method for diagnosing a battery using a battery diagnostic device, A step of obtaining state data including state values ​​of each of the factors for a plurality of batteries; A step of deriving a regression line representing the relationship between the factors using the above state data; A step of calculating the distance between the coordinates corresponding to the state values ​​of the above factors and the regression line; and A battery diagnosis method comprising the step of inputting diagnostic data including the above distance into a pre-trained diagnostic model, and determining an abnormal battery among the batteries using the diagnostic result output by the diagnostic model.

9. In Claim 8, The above factor is, A battery diagnostic method comprising one or more of the voltage, current, temperature, and capacity of the battery.

10. In Claim 8, The step of deriving the above regression line is, A battery diagnostic method comprising the step of performing a predefined linear regression analysis using state values ​​for each of the two factors to derive a regression line representing the relationship between the two factors.

11. In Claim 10, A battery diagnostic method comprising the step of calculating the distance, wherein the step of calculating the vertical distance between the two-dimensional coordinates corresponding to the state values ​​of each of the two factors and the regression line.

12. In Claim 10, The step of deriving the above regression line is, If the above factors are three or more, a step of deriving a plurality of combinations consisting of two factors; and A battery diagnostic method comprising the step of deriving a regression line for each of the above combinations.

13. In claim 8, The above diagnostic model is, A battery diagnostic method that is pre-trained to output diagnostic result data indicating whether a specific battery has a low voltage defect when diagnostic data for that specific battery is input.

14. In Claim 8, The above diagnostic data is, A battery diagnostic method comprising one or more of the average value of the distances for the plurality of batteries and the standard deviation value of the distances.

15. A data collection device for collecting state data, including state values ​​of each of the factors for a plurality of batteries; and A battery diagnostic device comprising acquiring state data from the data collection device and diagnosing whether the batteries are abnormal using the state data, The above battery diagnostic device is, A battery diagnostic system that derives a regression line representing the relationship between the factors using the above state data, calculates the distance between the coordinates corresponding to the state values ​​of the factors and the regression line, and determines an abnormal battery among the batteries using the calculated distance and a pre-trained diagnostic model.

Images