Battery diagnostic apparatus and operation method thereof

Abstract

A battery diagnostic apparatus, according to an embodiment disclosed in the present document, may comprise: an information acquisition unit for acquiring battery data for each charge / discharge cycle of each of a plurality of battery cells included in a battery pack; and a controller for calculating prediction data predicting whether each battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells on the basis of the battery data, preprocessing the prediction data in at least one manner, and deriving a diagnostic result for each charge / discharge cycle of each of the plurality of battery cells on the basis of the preprocessed prediction data.

Description

Battery diagnostic device and its operating method

[0001] Cross-citation with related applications

[0002] This invention claims the benefit of priority from Korean Patent Application No. 10-2024-0011496, filed January 25, 2024, the entire contents of which are incorporated herein by reference.

[0003] Technology field

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

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

[0006] These secondary batteries, in the form of battery packs (or modules) comprised of multiple cells, are increasingly being used in diverse fields, including electric vehicles. Consequently, the importance of battery pack diagnosis is growing. In particular, because battery packs contain numerous cells and secondary batteries can be charged and discharged, technology is needed to identify which cells are experiencing abnormalities and when they occur.

[0007] Conventional battery diagnostic methods have the problem of being unable to pinpoint the exact point (cycle) at which a battery cell malfunction occurs. Consequently, any cell experiencing a malfunction must be treated as having the malfunction at all points (cycles), leading to a high rate of over-diagnosis in battery pack diagnostic results.

[0008] One purpose of the embodiments disclosed in this document is to provide a battery diagnostic device and an operating method thereof that can more accurately identify the point in time when an abnormality occurs in a battery cell.

[0009] In addition, one purpose of the embodiments disclosed in this document is to provide a battery diagnosis device and an operating method thereof that can detect abnormal signs of a battery through diagnosis of a battery in actual operation and prevent accidents such as fire.

[0010] The technical problems of the embodiments disclosed in this document are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the descriptions below.

[0011] According to an embodiment disclosed in the present document, a battery diagnosis device may include an information acquisition unit that acquires battery data in charge / discharge cycles of each of a plurality of battery cells included in a battery pack, and a controller that, based on the battery data, produces prediction data predicting whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells, preprocesses the prediction data in at least one manner, and derives a diagnosis result for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data.

[0012] According to an embodiment, the controller can transform the prediction data by assigning a weight determined according to the charge / discharge cycle to the prediction data.

[0013] According to an embodiment, the controller may express the prediction data as binary data for each of the plurality of battery cells and convert the prediction data based on the number of charge / discharge cycles during which the prediction data has a first value.

[0014] According to an embodiment, the controller may, for each of the plurality of battery cells, if the number of charge / discharge cycles in which the prediction data has the first value is greater than or equal to a preset ratio, convert the prediction data for all charge / discharge cycles into the first value, and if the number of charge / discharge cycles in which the prediction data has the first value is less than the preset ratio, convert the prediction data for all charge / discharge cycles into a median value of the prediction data.

[0015] According to an embodiment, the controller may apply an interquartile range (IQR) to the prediction data to remove data that falls outside a preset range.

[0016] According to an embodiment, the controller can derive the diagnosis result by comparing the preprocessed prediction data with a threshold value.

[0017] According to an embodiment, the controller can obtain the prediction data using a pre-learned diagnostic model.

[0018] According to an embodiment disclosed in the present document, a battery diagnosis method may include a step of obtaining battery data for each charge / discharge cycle of a plurality of battery cells included in a battery pack, a step of generating prediction data predicting whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells based on the battery data, a step of preprocessing the prediction data in at least one manner, and a step of deriving a diagnosis result for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data.

[0019] According to an embodiment, the step of preprocessing the prediction data in at least one manner may be characterized by transforming the prediction data by assigning a weight determined according to the charge / discharge cycle to the prediction data.

[0020] According to an embodiment, the step of preprocessing the prediction data in at least one manner may be characterized by expressing the prediction data as binary data and converting the prediction data based on the number of charge / discharge cycles during which the prediction data has a first value.

[0021] According to an embodiment, the step of preprocessing the prediction data in at least one manner may be characterized in that, if the number of charge / discharge cycles having the prediction data having the first value is greater than or equal to a preset ratio, the prediction data for all charge / discharge cycles is converted into the first value, and if the number of charge / discharge cycles having the first value is less than the preset ratio, the prediction data for all charge / discharge cycles is converted into a median value of the prediction data.

[0022] According to an embodiment, the step of preprocessing the prediction data in at least one manner may be characterized by applying an interquartile range (IQR) to the prediction data to remove data outside a preset range.

[0023] The battery diagnostic device and its operating method according to the embodiments disclosed in this document can more accurately identify the point in time when an abnormality occurs in a battery cell. Accordingly, the over-inspection rate of battery diagnostic results can be reduced.

[0024] In addition, various effects may be provided, either directly or indirectly, through this document.

[0025] FIG. 1 is a block diagram showing the configuration of a battery diagnostic device according to one embodiment disclosed in this document.

[0026] Figures 2a and 2b are drawings showing examples of results obtained by applying a preprocessing method according to one embodiment disclosed in this document.

[0027] Figures 3a and 3b are drawings showing examples of results obtained by applying a preprocessing method according to another embodiment disclosed in this document.

[0028] Figure 3c is a diagram showing a process in which a preprocessing method according to another embodiment disclosed in this document is applied.

[0029] Figures 4a and 4b are drawings showing examples of results obtained by applying a preprocessing method according to another embodiment disclosed in this document.

[0030] FIG. 5 is a flowchart illustrating a battery diagnosis method according to one embodiment disclosed in this document.

[0031] FIG. 6 is a block diagram showing the hardware configuration of a computing system for performing an operating method of a battery diagnostic device according to one embodiment disclosed in this document.

[0032] Hereinafter, various embodiments of the present invention will be described with reference to the attached drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that the present invention encompasses various modifications, equivalents, and / or alternatives of the embodiments.

[0033] In this document, the singular form of a noun corresponding to an item may include one or more of said items, unless the context clearly indicates otherwise. In this document, phrases such as "A or B," "at least one of A and B," "at least one of A or B," "A, B, or C," "at least one of A, B, and C," and "at least one of A, B, or C" may each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish the corresponding element from other corresponding elements, and do not limit the corresponding elements in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as being “coupled” or “connected” to another component (e.g., a second component), with or without the terms “functionally” or “communicatively,” it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0034] Each component (e.g., a module or a program) described in this document may include one or more entities. According to various embodiments, one or more components or operations of the components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0035] The term "module" or "part" used in this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0036] Various embodiments of the present document may be implemented as software (e.g., a program or an application) including one or more instructions stored in a machine-readable storage medium (e.g., memory). For example, a processor of the device may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the device to operate to perform at least one function according to the at least one instruction called. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory" only means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0037] FIG. 1 is a block diagram showing the configuration of a battery diagnostic device according to one embodiment disclosed in this document.

[0038] Referring to FIG. 1, a battery diagnostic device (100) may include an information acquisition unit (110) and a controller (120).

[0039] A battery diagnostic device (100) can diagnose abnormalities in a battery pack containing multiple battery cells. In particular, the battery diagnostic device (100) can diagnose abnormalities in each battery cell for each charge / discharge cycle. Through this, the battery diagnostic device (100) can identify the point in time when an abnormality occurred in a battery cell, thereby increasing the diagnostic accuracy of the battery pack and reducing the over-inspection rate.

[0040] The operation of the battery diagnostic device (100) below can be performed by a battery management system (BMS) within a vehicle, a battery BMS provided within a battery pack, and can also be performed in various devices such as a server, cloud, charger, or charger / discharger.

[0041] The information acquisition unit (110) can acquire battery data from each charge / discharge cycle of a plurality of battery cells included in a battery pack. Here, each charge / discharge cycle may refer to a cycle in which charging and discharging of the battery are performed.

[0042] Battery data may include data regarding the battery's voltage. Additionally, the information acquisition unit (110) may also acquire additional battery-related data, such as the temperature, SOH, and SOC of each battery cell.

[0043] The information acquisition unit (110) may be implemented as a component related to a BMS (Battery Management System), a charger / discharger, or a battery. In some cases, the information acquisition unit (110) may be implemented as a separate component (e.g., an external server) from the BMS and the charger / discharger to acquire data from the BMS, the charger / discharger, etc.

[0044] The controller (120) can generate prediction data predicting whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells based on battery data. The controller (120) can primarily predict whether a battery cell is abnormal for each charge / discharge cycle.

[0045] According to an embodiment, the controller (120) can generate prediction data using a pre-trained diagnostic model. The diagnostic model can be trained to derive diagnostic results using training data. For example, the diagnostic model can be trained using a supervised learning method that receives training data pairs (battery data - abnormality) as input and predicts diagnostic results. As another example, the diagnostic model can be trained using an unsupervised learning method that receives only battery data as input and predicts diagnostic results. In addition, the diagnostic model can of course be trained using various learning methods.

[0046] The output of a pre-trained diagnostic model can be in the form of binary data indicating whether a battery cell is abnormal. For example, a pre-trained diagnostic model may output 1 if the battery cell is predicted to be abnormal and 0 if the battery cell is predicted to be normal. In another example, a pre-trained diagnostic model may output the probability that a battery cell is abnormal.

[0047] The controller (120) can preprocess the prediction data in at least one manner. By preprocessing the prediction data, the controller (120) can further improve the diagnostic accuracy.

[0048] In one embodiment, the controller (120) can preprocess the prediction data by assigning weights. The controller (120) can transform the prediction data by assigning weights determined based on the charge / discharge cycle. In other words, the controller (120) can preprocess the prediction data by calculating the weights with the prediction data.

[0049] The weight may be determined according to the charge / discharge cycle. For example, the controller (120) may assign a weight that increases as the charge / discharge cycle progresses. This is because the battery cell initially has a low abnormality occurrence rate, and the abnormality occurrence rate may increase as the number of charge / discharge cycles increases. For example, if the number of charge / discharge cycles is n (where n is a natural number greater than or equal to 2), the weight assigned to the kth cycle (where k is a natural number less than or equal to n) may be k / n. The above-described weights are merely examples, and the method of assigning weights is not limited.

[0050] According to another embodiment, the controller (120) can express the prediction data as binary data.

[0051] The controller (120) may express the prediction data as binary data. For example, the binary data may refer to data having either a first value (e.g., 1) or a second value (e.g., 0). Here, the first value may indicate that the battery cell is abnormal, and the second value may indicate that the battery cell is normal.

[0052] Predictive data can be initially expressed as binary data, and in some cases, data expressed as probability values ​​can be converted into binary data. In this case, the predicted data can be converted into binary data by comparing the probability value of the predicted data with a reference value (e.g., 0.5). For example, if the probability value of the predicted data is greater than or equal to the reference value, the predicted data can be converted to a first value (1), and if the probability value of the predicted data is less than the reference value, the predicted data can be converted to a second value (0).

[0053] According to an embodiment, the controller (120) can convert the prediction data based on the number of charge / discharge cycles in which the prediction data has the first value. That is, the controller (120) can convert the prediction data based on the distribution of the prediction data according to the cycle.

[0054] For example, if the number of charge / discharge cycles in which the predicted data has the first value is greater than or equal to a preset ratio, the controller (120) may convert the predicted data for all charge / discharge cycles to the first value. If the predicted data has the first value in a preset ratio or more of all cycles, the controller (120) may convert the predicted data for all cycles to the first value to reduce prediction inaccuracy, since it means that the period in which the corresponding battery cell is predicted to be abnormal is long. The preset ratio may be set to, for example, 20%.

[0055] The controller (120) can convert the prediction data for all charge / discharge cycles into a median value of the prediction data if the number of charge / discharge cycles in which the prediction data has the first value is less than a preset ratio. That is, the controller (120) can convert the prediction data for all cycles into a value that has a greater number of cycles between the first value and the second value.

[0056] The controller (120) can reduce the inaccuracy of abnormal prediction by converting the prediction data in this way.

[0057] According to another embodiment, the controller (120) can preprocess the prediction data by removing a portion of the prediction data.

[0058] The controller (120) can apply an interquartile range (IQR) to the prediction data to remove data that falls outside a preset range. Through this, the controller (120) can remove outliers (noise) from the prediction data, thereby improving diagnostic accuracy.

[0059] The interquartile range is a type of method for analyzing the distribution of multiple data groups. It can mean a method of dividing the distribution of data into four ranges and excluding the 1 / 4 ranges at both ends as outliers. At this time, the value obtained by subtracting the 1st quartile (25% percentile, Q1) from the 3rd quartile (75% percentile, Q3) of the data distribution can be used as the standard (IQR = Q3 - Q1). For example, the controller (120) can remove data that fall within a range exceeding Q3 + 1.5 * (Q3 - Q1) and a range less than Q1 - 1.5 * (Q3 - Q1) from the distribution of predicted data as outliers.

[0060] According to an embodiment, the controller (120) can derive cycle-by-cycle diagnostic results for each of a plurality of battery cells based on preprocessed prediction data.

[0061] In an embodiment, the controller (120) may compare preprocessed prediction data with a threshold value to derive a diagnostic result. The threshold value may be preset, for example, 0.5.

[0062] For example, if the preprocessed prediction data for a specific cycle of a specific cell is below a threshold value, the controller (120) may determine the diagnostic result for that cycle as normal. Conversely, if the preprocessed prediction data for a specific cycle of a specific cell is above a threshold value, the controller (120) may determine the diagnostic result for that cycle as abnormal.

[0063] In addition, if the diagnosis results confirm that the battery cell is defective, the controller (120) can provide information about the defective battery cell to the user. For example, the controller (120) can provide information about the defective battery cell to the user terminal via a communication unit (not shown), and can also provide information about the defective battery cell via a display provided in a vehicle or charger.

[0064] In this way, the battery diagnostic device (100) can perform abnormal diagnosis for each cycle of each cell, thereby confirming the point in time when an abnormality occurred in an abnormal battery cell, and can increase the diagnostic accuracy and reduce the over-inspection rate for the entire battery pack.

[0065] Figures 2a and 2b are drawings showing examples of results obtained by applying a preprocessing method according to one embodiment disclosed in this document.

[0066] First, referring to FIG. 2a, the controller (120) can preprocess the prediction data by assigning weights to the prediction data, and an example (210) of the preprocessed prediction data is illustrated.

[0067] The controller (120) can produce predicted data (Predicted) for each cycle (CycNumber) of a specific cell (QueryID) and can assign a weight (weighted) according to the cycle to the predicted data. The controller (120) can preprocess the predicted data (weighted_predicted) by calculating the predicted data and the weight. In Fig. 2a, the product (or inner product) of the predicted data and the weight is illustrated as an example, but it is obvious that various calculations can be applied to the calculation of the weight.

[0068] FIG. 2b illustrates an example of a graph (230) showing the results of preprocessing predicted data (220) by assigning weights to the controller (120). In FIG. 2b, the x-axis represents cycles and the y-axis represents data values. The controller (120) can lower the over-diagnosis rate by preprocessing predicted data by assigning weights according to cycles. Referring to FIG. 2b, it can be confirmed that the number of cycles judged as abnormal is reduced as weights are assigned.

[0069] Figures 3a and 3b are drawings showing examples of results obtained by applying a preprocessing method according to another embodiment disclosed in this document.

[0070] Referring to FIG. 3a, the controller (120) can preprocess data by transforming the values ​​of the data according to the data distribution of the prediction data, and an example (310) of the preprocessed prediction data is illustrated.

[0071] For example, in FIG. 3a, the controller (120) can convert prediction data for all cycles into the first value since the number of cycles (10) having the first value (1) is greater than or equal to a preset ratio (0.2) of the total number of cycles (18). In this way, the prediction data (330) converted by preprocessing the prediction data (320) can be expressed as in FIG. 3b.

[0072] Figure 3c is a diagram showing a process in which a preprocessing method according to another embodiment disclosed in this document is applied.

[0073] Referring to FIG. 3c, the controller (120) can convert the prediction data according to the number of cycles in which the prediction data is the first value.

[0074] First, the controller (120) can check whether the number of cycles in which the predicted data is the first value is greater than or equal to a preset ratio (340).

[0075] The controller (120) can convert the prediction data of all cycles into the first value if the number of cycles in which the prediction data is the first value is greater than or equal to a preset ratio (Yes) (350). Conversely, the controller (120) can convert the prediction data of all cycles into the median value of the prediction data if the number of cycles in which the prediction data is the first value is less than a preset ratio (No) (360).

[0076] Figures 4a and 4b are drawings showing examples of results obtained by applying a preprocessing method according to another embodiment disclosed in this document.

[0077] Referring to FIG. 4a, the controller (120) can preprocess a portion of the prediction data by removing outliers, and an example (410) of the preprocessed prediction data is illustrated. In FIG. 4a, it can be confirmed that the prediction data of the first and fourth cycles have been removed.

[0078] The controller (120) can apply an interquartile range to the predicted data to remove data that falls outside the preset range. Data located above the dotted line in FIG. 4b represent data that falls outside the preset range, and thus can be removed.

[0079] FIG. 5 is a flowchart illustrating a battery diagnosis method according to one embodiment disclosed in this document.

[0080] Referring to FIG. 5, a battery diagnosis method may include a step (S100) of obtaining battery data for each charge / discharge cycle of a plurality of battery cells included in a battery pack, a step (S200) of generating prediction data predicting whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells based on the battery data, a step (S300) of preprocessing the prediction data in at least one manner, and a step (S400) of deriving a diagnosis result for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data.

[0081] In step S100, the information acquisition unit (110) can acquire battery data for each charge / discharge cycle for each battery cell. The battery data can include voltage data.

[0082] At step S200, the controller (120) can generate prediction data for each charge / discharge cycle for each battery cell. In an embodiment, the controller (120) can obtain prediction data using a pre-learned diagnostic model, and the prediction data can refer to data predicting whether a specific cell has an abnormality in a specific cycle.

[0083] At step S300, the controller (120) may preprocess the prediction data in at least one manner. In one embodiment, the controller (120) may preprocess the prediction data by assigning a weight determined according to the cycle to the prediction data. In another embodiment, the controller (120) may preprocess the prediction data based on the distribution of prediction data values ​​according to the cycle. In yet another embodiment, the controller (120) may preprocess the prediction data by removing data based on the range of the prediction data.

[0084] At step S400, the controller (120) can derive diagnostic results for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data. In an embodiment, the controller (120) can derive diagnostic results by comparing the preprocessed prediction data with a threshold value. For example, if the preprocessed prediction data for a specific cell in a specific cycle is greater than or equal to the threshold value, the controller (120) can diagnose the cell as abnormal in the specific cycle.

[0085] FIG. 6 is a block diagram showing the hardware configuration of a computing system for performing an operating method of a battery diagnostic device according to one embodiment disclosed in this document.

[0086] Referring to FIG. 6, a computing system (1000) according to one embodiment disclosed in the present document may include an MCU (1010), a memory (1020), an input / output I / F (1030), and a communication I / F (1040).

[0087] The MCU (1010) may be a processor that executes various programs stored in the memory (1020), processes various information including time series data of the battery through these programs, and performs the functions of the controller included in the battery diagnostic device shown in the aforementioned FIG. 1.

[0088] The memory (1020) can store various programs for performing the functions of the battery diagnostic device. In addition, the memory (1020) can store various information, including battery data, diagnostic prediction results, etc., and can store a previously learned diagnostic model.

[0089] Such memories (1020) may be provided in multiples as needed. The memories (1020) may be volatile memories or non-volatile memories. As volatile memories (1020), RAM, DRAM, SRAM, etc. may be used. As non-volatile memories (1020), ROM, PROM, EAROM, EPROM, EEPROM, flash memories, etc. may be used. The examples of the memories (1020) listed above are merely examples and are not limited to these examples.

[0090] The input / output I / F (1030) can provide an interface that enables data transmission and reception between an input device (not shown) such as a keyboard, mouse, or touch panel, and an output device (not shown) such as a display and the MCU (1010).

[0091] The communication I / F (1040) is a component capable of transmitting and receiving various data with a server, and may be any device capable of supporting wired or wireless communication. For example, a battery diagnostic device can transmit and receive various information, including battery time-series data, from a separately provided external server via the communication I / F (1040).

[0092] In this way, a computer program according to one embodiment disclosed in this document may be implemented as a module that is recorded in a memory (1020) and processed by an MCU (1010) to perform each function illustrated in FIG. 2, for example.

[0093] Although all components constituting the embodiments disclosed in this document have been described as being combined or operating in combination as one, the embodiments disclosed in this document are not necessarily limited to such embodiments. That is, within the scope of the purpose of the embodiments disclosed in this document, all of the components may be selectively combined and operated one or more times.

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

[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.

Claims

1. An information acquisition unit that acquires battery data in each charge / discharge cycle of a plurality of battery cells included in a battery pack; and Based on the above battery data, prediction data is generated to predict whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells, Preprocessing the above prediction data in at least one way, A battery diagnostic device comprising a controller that derives diagnostic results for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data.

2. In paragraph 1, The above controller, A battery diagnostic device that converts the prediction data by assigning a weight determined according to the charge / discharge cycle to the prediction data.

3. In paragraph 1, The above controller, For each of the above plurality of battery cells, Express the above prediction data as binary data, A battery diagnostic device that converts the prediction data based on the number of charge / discharge cycles in which the prediction data has a first value.

4. In paragraph 3, The above controller, For each of the above plurality of battery cells, If the number of charge / discharge cycles in which the predicted data has the first value is greater than or equal to a preset ratio, the predicted data for all charge / discharge cycles are converted to the first value, A battery diagnostic device that converts predicted data for all charge / discharge cycles into a median value of the predicted data when the number of charge / discharge cycles having the first value is less than the preset ratio.

5. In paragraph 1, The above controller, A battery diagnostic device that removes data outside a preset range by applying an interquartile range (IQR) to the above prediction data.

6. In paragraph 1, The above controller, A battery diagnosis device that compares the preprocessed prediction data with a threshold value to derive the diagnosis result.

7. In paragraph 1, The above controller, A battery diagnostic device that obtains the above prediction data using a learned diagnostic model.

8. A step of acquiring battery data in each charge / discharge cycle of a plurality of battery cells included in a battery pack; A step of generating prediction data predicting whether a battery cell is abnormal for each charge / discharge cycle of each of the plurality of battery cells based on the battery data; a step of preprocessing the above prediction data in at least one manner; and A battery diagnosis method, comprising a step of deriving a diagnosis result for each charge / discharge cycle of each of the plurality of battery cells based on the preprocessed prediction data.

9. In paragraph 8, The step of preprocessing the above prediction data in at least one way comprises: A battery diagnosis method characterized in that the prediction data is converted by assigning a weight determined according to the charge / discharge cycle to the prediction data.

10. In paragraph 8, The step of preprocessing the above prediction data in at least one way comprises: A battery diagnosis method characterized in that the prediction data is expressed as binary data and the prediction data is converted based on the number of charge / discharge cycles in which the prediction data has a first value.

11. In paragraph 10, The step of preprocessing the above prediction data in at least one way comprises: If the number of charge / discharge cycles in which the predicted data has the first value is greater than or equal to a preset ratio, the predicted data for all charge / discharge cycles are converted to the first value, A battery diagnosis method characterized in that, when the number of charge / discharge cycles having the first value is less than the preset ratio, the predicted data for all charge / discharge cycles is converted into a median value of the predicted data.

12. In paragraph 8, The step of preprocessing the above prediction data in at least one way comprises: A battery diagnosis method characterized in that data outside a preset range are removed by applying an interquartile range (IQR) to the above prediction data.

Images