Battery management apparatus and method

Abstract

According to an embodiment disclosed in this document, a battery management device may include a memory and a control unit, and the control unit may be configured to perform a first number of charge-discharge cycles on a first battery cell to obtain first charge-discharge data, and input data including the first charge-discharge data to a 2D CNN (Convolutional Neural Network) trained based on second charge-discharge data obtained by performing a second number of charge-discharge cycles on a second battery cell, and predict the State of Health (SOH) of the first battery cell based on result data output through the 2D CNN in response to the input data.

Description

Technology Field

[0001] The embodiments disclosed in this document relate to a battery management device and method. Background Technology

[0002] With the proliferation of electric vehicles powered by electricity, research and development on new vehicle architectures is actively underway. For example, electric vehicles can be powered by rechargeable batteries; here, rechargeable batteries refer to rechargeable batteries that include conventional Ni / Cd and Ni / MH batteries, as well as recent lithium-ion batteries. Among rechargeable batteries, lithium-ion batteries have the advantage of significantly higher energy density compared to conventional Ni / Cd and Ni / MH batteries. Furthermore, lithium-ion batteries can be manufactured in a compact and lightweight manner, making them suitable for use as power sources for mobile devices. Recently, their scope of application has expanded to include power sources for electric vehicles, drawing attention as a next-generation energy storage medium.

[0003] Such battery cells, battery modules, battery packs, or battery racks can be utilized in various devices. For example, batteries can be used in mobile devices such as mobile phones, laptop computers, smartphones, and smart pads, as well as in fields such as electric vehicles (EVs, HEVs, PHEVs) and large-capacity energy storage systems (ESS).

[0004] The status and operation of these batteries can be managed and controlled by a battery management system (BMS). The battery management system can be included together with the batteries within a single device.

[0005] Meanwhile, battery cells can be charged and discharged based on various charge-discharge profiles. Through charge-discharge cycles of the battery cells, the degree of degradation or degradation patterns over time can be identified. Based on the identification results, the replacement cycle or stability of the battery cells can be determined, and the State of Health (SOH) of other battery cells can be predicted based on the identified degree of degradation or degradation patterns.

[0006] The background description provided herein is intended to provide a general context for the present disclosure. Unless otherwise specified in this specification, the materials described in this section are not prior art for the claims of this application and are not recognized as prior art or proposals of prior art by being included in this section. [Prior Art Documents] (Patent Document 0001) Published Patent Application No. 10-2020-0119383 (Patent Document 0002) Published Patent Application No. 10-2020-0140093 (Patent Document 0003) Published Patent Application No. 10-2013-0113296 The problem to be solved

[0007] It is necessary to accurately identify the replacement cycle of battery cells while ensuring the safety of the devices (e.g., electric vehicles) in which the battery cells are used, by predicting and analyzing the degradation process of battery cells as they charge and discharge at an early stage.

[0008] One objective of the embodiments disclosed in this document is to provide a battery management device that accurately and efficiently predicts the State of Health (SOH) of a battery cell based on the analysis of only charge / discharge data obtained from a small number of charge / discharge cycles, based on a learned artificial intelligence model.

[0009] The technical problems of the embodiments disclosed in this document are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0010] According to an embodiment disclosed in this document, a battery management device may include a memory for storing at least one instruction and a control unit operatively connected to said memory. For example, said at least one instruction may be configured such that, when executed by said control unit, a first number of charge-discharge cycles are performed on a first battery cell to obtain first charge-discharge data, and said first charge-discharge data is input to a 2D CNN (Convolutional Neural Network) trained based on second charge-discharge data obtained by performing a second number of charge-discharge cycles on at least one second battery cell, and said first charge-discharge data is predicted based at least partially on result data output through said 2D CNN in response to said first charge-discharge data.

[0011] According to an embodiment, the first number may be characterized as being less than the second number.

[0012] According to an embodiment, the first charge / discharge data includes one or more data sets regarding voltage data, current data, and temperature data of the first battery cell corresponding to each of the charge / discharge cycles, and each of the one or more data sets may correspond to an instance of the charge / discharge cycle performed in the first battery cell.

[0013] According to an embodiment, when the at least one instruction is executed by the control unit, the battery management device may be configured to identify the number of data sets included in the first charge / discharge data during a specified period, and in response to determining that the number of data sets is smaller than the first number corresponding to the first number, determine that at least a part of the first charge / discharge data is missing or incomplete, determine a target number of additional cycle repetitions by calculating the difference between the first number and the number of data sets, and perform additional charge / discharge cycles to obtain third charge / discharge data.

[0014] According to an embodiment, when the at least one instruction is executed by the control unit, the battery management device may be configured to input the third charge / discharge data into the 2D CNN, generate the result data based at least partially on the third charge / discharge data, and predict the State of Health (SOH) of the first battery cell based on the result data.

[0015] According to an embodiment, the target number of additional cycle repetitions may be characterized as being less than or equal to the first number.

[0016] According to an embodiment, when the at least one instruction is executed by the control unit, the battery management device compares the actual SOH of the first battery cell with the prediction result generated by the 2D CNN based on the first charge / discharge data, identifies error data based on the comparison result between the prediction result and the actual SOH, and the error data includes at least one of RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), STD (Standard Deviation), or any combination thereof; and can be configured to update the number of target charge / discharge cycles for SOH prediction or further train the 2D CNN using the error data.

[0017] According to an embodiment disclosed in this document, a battery management method may include the steps of: a control unit performing a first number of charge-discharge cycles for a first battery cell to obtain first charge-discharge data; the control unit inputting the first charge-discharge data to a 2D CNN (Convolutional Neural Network) trained based on second charge-discharge data obtained by performing a second number of charge-discharge cycles for at least one second battery cell; and the control unit predicting the State of Health (SOH) of the first battery cell based at least partially on result data output through the 2D CNN in response to the first charge-discharge data.

[0018] According to an embodiment, the battery management method may further include the steps of: the control unit identifying the number of data sets included in the first charge / discharge data during a specified period; the control unit determining that at least a portion of the first charge / discharge data is missing or incomplete in response to the determination that the number of the identified data sets is smaller than a first number corresponding to the first number; determining a target number of additional cycle repetitions by calculating the difference between the first number and the number of data sets; and obtaining third charge / discharge data by performing additional charge / discharge cycles according to the target number of additional cycle repetitions.

[0019] According to an embodiment, the battery management method may further include the steps of: the control unit inputting the third charge / discharge data into the 2D CNN; generating the result data based at least partially on the third charge / discharge data; and the control unit predicting the State of Health (SOH) of the first battery cell based on the result data.

[0020] According to an embodiment, the battery management method may further include the steps of: the control unit comparing a prediction result generated by the 2D CNN based on the first charge / discharge data with the actual SOH of the first battery cell; identifying error data based on the comparison result between the prediction result and the actual SOH; the error data including at least one of RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), STD (Standard Deviation), or any combination thereof; and using the error data to update the number of target charge / discharge cycles for SOH prediction or further training the 2D CNN. Effects of the invention

[0021] The battery management device and method according to the embodiments disclosed in this document can provide an algorithm that stably and efficiently predicts the SOH of a battery with only a small number of input data under various charge / discharge profiles.

[0022] In addition, various effects that can be identified directly or indirectly through this document may be provided. Brief explanation of the drawing

[0023] The attached drawings illustrate preferred embodiments of the present disclosure and are intended to provide a further understanding of the technical features of the present disclosure together with the aforementioned disclosure; therefore, the present disclosure is not to be interpreted as being limited to the drawings. FIG. 1 is a block diagram showing the configuration of a battery management device according to one embodiment disclosed in this document. FIG. 2 is an exemplary block diagram showing the structure of a 2D CNN according to one embodiment disclosed in this document. FIG. 3a is a graph of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document. FIG. 3b is a graph of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document. FIG. 3c is a graph of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document. FIG. 3d is a graph of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document. FIG. 4 is a table of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document. FIG. 5 is a flowchart of a battery management method according to one embodiment disclosed in this document. FIG. 6 is a block diagram showing the hardware configuration of a computing system for performing a method of operation of a battery management device according to one embodiment disclosed in this document. Specific details for implementing the invention

[0024] Hereinafter, various embodiments of the present invention are described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments and should be understood to include various modifications, equivalents, and / or alternatives of the embodiments of the present invention.

[0025] In this document, the singular form of a noun corresponding to an item may include one or more of said items unless the relevant 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 the corresponding phrase, or all possible combinations thereof. Terms such as “first,” “second,” or “first” or “second” may be used simply to distinguish a component from another corresponding component and do not limit the components in any other aspect (e.g., importance or order). Where any (e.g., 1st) component is referred to as “coupled” or “connected” to another (e.g., 2nd) component, with or without the terms “functionally” or “communicationly,” it means that said any component may be connected to said other component directly (e.g., via a wire), wirelessly, or through a third component.

[0026] Each component (e.g., module or program) of the components described in this document may include a singular or multiple entities. According to various embodiments, one or more of the components or operations may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or 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 multiple components in the same or similar manner as they were performed by the corresponding component among the multiple components prior to the integration. According to various embodiments, 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.

[0027] As used in this document, the terms "module" or "...part" 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, for example. A module may be a component formed integrally, or a minimum unit of said component or a part thereof 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).

[0028] Various embodiments of this document may be implemented as software (e.g., a program or application) comprising one or more instructions stored in a storage medium (e.g., memory) readable by a machine. For example, the processor of the machine may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the machine to operate to perform at least one function according to the at least one called instruction. The one or more instructions may include code generated by a compiler or code that can be executed by an interpreter. The storage medium readable by a machine may be provided in the form of a non-transitory storage medium. Here, "non-transitory" simply 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.

[0030] FIG. 1 is a block diagram showing the configuration of a battery management device according to one embodiment disclosed in this document.

[0031] Referring to FIG. 1, the battery management device (100) may include a memory (110) and / or a control unit (120).

[0032] According to one embodiment, the memory (110) may store commands or data. For example, the memory (110) may store one or more instructions that cause the battery management device (100) to perform various operations when executed by the control unit (120).

[0033] For example, the memory (110) may be implemented as a single chipset with the control unit (120). The control unit (120) may include at least one of a communication processor or a modem.

[0034] For example, the memory (110) can store various information related to the battery management device (100). For example, the memory (110) can store information regarding the operation history of the control unit (120). For example, the memory (110) can store information related to the status and / or operation of other components of the battery management device (100) (e.g., sensor unit, display unit, interface, battery cell, or at least one of any combination thereof).

[0035] For example, the memory (110) may include multiple storage devices of different types. For example, the memory (110) may include at least one of RAM (random-access memory), eMMC (Embedded Multi-Media Card), or any combination thereof.

[0036] According to one embodiment, the control unit (120) may be operatively connected to the memory (120). For example, the control unit (120) may control the operation of the memory (110).

[0037] For example, the control unit (120) can obtain first charge / discharge data by performing a first number of charge / discharge cycles on the first battery cell.

[0038] For example, charging and discharging may be performed based on a specified charge-discharge profile for a first battery cell. The specified charge-discharge profile may include, for example, setting values ​​regarding the magnitude of pressure on the battery cell, ambient temperature, and charging mode (e.g., fast charging, normal charging, slow charging). The charging mode may be classified according to, for example, a charging pattern (e.g., QC (Quick Charge) or SC (Slow Charge)), voltage and / or current conditions for charging termination, rest time after charging, a discharge pattern, voltage and / or current conditions for discharging termination, and rest time after discharging.

[0039] For example, the control unit (120) can charge and discharge the first battery cell a first number of times based on a predefined charge / discharge profile. The process of charging and discharging the battery cell can be defined as one charge / discharge cycle.

[0040] For example, the control unit (120) may input or provide input data including first charge / discharge data to a 2D CNN (Convolutional Neural Network).

[0041] For example, a 2D CNN may include an artificial intelligence model trained based on second charge-discharge data obtained by performing a second number of charge-discharge cycles on a second battery cell.

[0042] For example, the control unit (120) can train a 2D CNN using second charge / discharge data obtained while performing a second number of charge / discharge cycles (e.g., 270 times) for the second battery cell.

[0043] For example, a 2D CNN may include an input layer, a pooling layer, a non-pooling layer, a linear layer, and an output layer. A 2D CNN may include as many 2D CNN models as there are pooling layers and non-pooling layers.

[0044] For example, the control unit (120) can determine the number of charge / discharge cycles to obtain an optimal prediction result by fixing the number of pooling layers to n (e.g., 1) and changing the number of non-pooling layers.

[0045] For example, the control unit (120) can predict the State of Health (SOH) of the first battery cell based on the result data output through a 2D CNN in response to the input data.

[0046] For example, the first time may be less than the second time.

[0047] For example, the first charge / discharge data and / or the second charge / discharge data may include at least one data set regarding the voltage, current, and temperature of a battery cell corresponding to each charge / discharge cycle. The first charge / discharge data may include, for example, a first number of data sets (e.g., 3 to 5) corresponding to each of a first number of charge / discharge cycles (e.g., 3 to 10 times). The second charge / discharge data may include, for example, a second number of data sets (e.g., 250 to 290) corresponding to each of a second number of charge / discharge cycles (e.g., 250 to 290 times).

[0048] For example, the control unit (120) can monitor whether the number of data sets included in the first charge / discharge data is less than the first number based on a specified period. The control unit (120) may determine, for example, that at least part of the first charge / discharge profile is missing if the number of data sets included in the first charge / discharge data is less than the first number corresponding to the first number. In this case, the control unit (120) may obtain the third charge / discharge data by performing additional charge / discharge cycles for the first battery cell by a target number of times corresponding to the difference between the first number and the number of data sets. The target number of times may be, for example, less than or equal to the first number.

[0049] For example, the control unit (120) can predict the SOH of the first battery cell based on the output result data by further inputting the third charge / discharge data into the 2D CNN. That is, if the control unit (120) identifies that the number of data sets included in the first charge / discharge data is less than the first number corresponding to the first number of times, it can acquire additional data sets by performing additional charge / discharge cycles and input the first number of data sets into the 2D CNN.

[0050] For example, the control unit (120) can update the first number of times (or the number of charge / discharge cycles for SOH prediction) based on the accuracy of the SOH prediction result of the first battery cell.

[0051] For example, the control unit (120) can input the first charge / discharge data into a 2D CNN and compare the output prediction result with the actual SOH of the first battery cell.

[0052] For example, the control unit (120) can identify error data including at least one of RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), STD (Standard Deviation), or any combination thereof, based on the comparison result between the prediction result and the actual SOH.

[0053] For example, the control unit (120) can update the number of target charge / discharge cycles for SOH prediction using error data. The control unit (120) can, for example, increase or decrease the number of target charge / discharge cycles based on error data. The number of target charge / discharge cycles may be 3 to 10 times, but this is exemplary and the embodiments of the present invention are not limited thereto.

[0054] For example, when the control unit (120) predicts the SOH of the third battery cell after the SOH prediction for the first battery cell is completed, it can obtain charge / discharge data for SOH prediction by performing charge / discharge cycles for the third battery cell as many times as the target number of charge / discharge cycles.

[0056] FIG. 2 is an exemplary block diagram showing the structure of a 2D CNN according to one embodiment disclosed in this document.

[0057] According to one embodiment, a battery management device (e.g., the battery management device (100) of FIG. 1) can predict the SOH of a battery cell based on a 2D CNN according to FIG. 2.

[0058] For example, a 2D CNN may include an input layer. A battery management device may input at least one charge / discharge data to the input layer of the 2D CNN.

[0059] For example, a 2D CNN may include pooling blocks that include at least one pooling layer.

[0060] For example, the battery management device is n p Input data can be processed through pooling layers. The battery management device, for example, n p Input data can be processed by fixing to 1. The pooling layer may include, for example, a 2D CNN layer, a batch normalization layer (Batchnorm), a Rectified Linear Unit (ReLu), and a Max Pooling layer.

[0061] For example, a 2D CNN may include non-pooling blocks that include at least one non-pooling layer.

[0062] For example, the battery management device is n npThrough non-pooling layers, data transmitted from the pooling block can be processed. The battery management unit, for example, to derive an optimal result n np Data can be processed by adjusting the number of layers within a specified range (e.g., 3 to 10). Non-pooling layers may include, for example, 2D CNN layers, batch normalization layers (Batchnorm), and ReLu (Rectified Linear Unit).

[0063] For example, a 2D CNN can include a linear layer.

[0064] For example, a battery management device can perform linear transformations on data transmitted from a non-pooling block through a linear layer.

[0065] For example, a 2D CNN can include an output layer.

[0066] For example, a battery management device can predict the SOH of a battery cell through result data output from an output layer.

[0068] In the following description of FIGS. 3a to 3d, the Mean Absolute Error (MAE) and Standard Deviation (STD) of the prediction results according to the type and number of charge-discharge cycles and the number of non-pooling layers will be discussed. For example, the x-axis of the graph represents the number of non-pooling layers, n. np And, the y-axis corresponds to the MAE and STD of the prediction results.

[0069] FIGS. 3a and 3b are graphs relating to result data output based on a specific charge / discharge cycle according to an embodiment disclosed in this document.

[0070] Reference numbers 301 and 302 represent the performance of a predicted SOH prediction result based on multiple charge-discharge data, including charge-discharge data obtained through the first charge-discharge cycle for a battery cell. That is, it represents the performance of a predicted SOH prediction result including charge-discharge data corresponding to the first charge-discharge cycle.

[0071] For example, n np If α is 1 and the charge / discharge cycle is 3 times, the MAE of the predicted result may be 3%.

[0072] For example, n np If α is 2 and the charge / discharge cycle is 3 times, the MAE of the predicted result may be about 2%.

[0073] For example, n np When α is 3 to 6 and the charge / discharge cycle is 3 to 90 times, the MAE of the prediction result may be less than 2%.

[0074] For example, n np If α is 1 and the charge / discharge cycle is 3 times, the STD of the predicted result may be 3%.

[0075] For example, n np If α is 2 and the charge / discharge cycle is 3 times, the STD of the predicted result may be about 1.7%.

[0076] For example, n np When the value is 3 to 6 and the charge / discharge cycle is 3 to 90 times, the STD of the predicted result may be less than 1.5%.

[0078] FIGS. 3c and FIGS. 3d are graphs relating to result data output based on a specific charge / discharge cycle according to an embodiment disclosed in this document.

[0079] Reference numbers 303 and 304 represent the performance of a SOH prediction result based on multiple charge-discharge data in which charge-discharge data obtained through the first charge-discharge cycle for the battery cell is missing. That is, it represents the performance of the SOH prediction result in a situation where charge-discharge data corresponding to the first charge-discharge cycle is missing.

[0080] For example, n np When α is 1 and the charge / discharge cycle is 3 times, the MAE of the predicted result may be approximately 2.8%.

[0081] For example, n np When α is 2 and the charge / discharge cycle is 3 times, the MAE of the predicted result may be approximately 1.9%.

[0082] For example, n np When α is 3 to 6 and the charge / discharge cycle is 3 to 90 times, the MAE of the prediction result may be less than 2%.

[0083] For example, n np If α is 1 and the charge / discharge cycle is 3 times, the STD of the predicted result may be about 1.9%.

[0084] For example, n np If α is 2 and the charge / discharge cycle is 3 times, the STD of the predicted result may be about 1.4%.

[0085] For example, n np When the value is 3 to 6 and the charge / discharge cycle is 3 to 90 times, the STD of the predicted result may be less than 1.6%.

[0086] For example, the battery management device n when the number of charge / discharge cycles is 5, 10, 30, or 90. np Even if it is small, it can be confirmed that MAE and STD are generally low.

[0087] Therefore, the battery management device n even when the number of charge / discharge cycles is low npWhen set to 3 to 6, it is confirmed that MAE and STD are generally low, and by using 2D CNN within that range, it is possible to output SOH prediction results with high accuracy through relatively low charge / discharge cycles.

[0088] Furthermore, since there is no significant difference in the accuracy of the prediction result even if charge / discharge data corresponding to a specific cycle (e.g., the first cycle) is missing, the battery management device can output the prediction result as is or output a SOH prediction result with robust accuracy by utilizing additional charge / discharge data corresponding to other cycles.

[0090] FIG. 4 is a table of result data output based on a specific charge / discharge cycle according to one embodiment disclosed in this document.

[0091] According to one embodiment, a battery management device (e.g., the battery management device (100) of FIG. 1) has a charge / discharge cycle count n cy , number of non-pooling layers n np The MAE and STD of the prediction results can be identified by changing the value. In addition, the battery management device can identify the MAE and STD of the prediction results in a situation where charge / discharge data corresponding to a specific cycle (e.g., the first cycle) is missing.

[0092] Referring to Figure 4, a prediction that includes charge / discharge data corresponding to a specific cycle is defined as Case 1, and a prediction that lacks charge / discharge data corresponding to a specific cycle can be defined as Case 2.

[0093] For example, in Case 1, the number of charge / discharge cycles n cy α is 5 to 10 and the number of non-pooling layers is n npWhen there are 4 or 6, the MAE and STD are low, so it can be confirmed that the accuracy of the prediction results is relatively high in this case. Based on this, the battery management device can update the target number of charge / discharge cycles or train a 2D CNN.

[0094] For example, in Case 2, the number of charge / discharge cycles n cy g is 5 and the number of non-pooling layers n np Since the MAE and STD are low when there are 3, it can be confirmed that the accuracy of the prediction results is relatively high in this case. Based on this, the battery management device can update the target number of charge / discharge cycles or train a 2D CNN.

[0095] In addition, according to the results of Case 1 and Case 2, the accuracy of the prediction result does not change significantly even if charge / discharge data corresponding to a specific cycle is missing. Therefore, even if some charge / discharge data is missing, the prediction result can be output as is, or a robust accuracy SOH prediction result can be output based on the charge / discharge data obtained by performing more charge / discharge cycles.

[0097] FIG. 5 is a flowchart of a battery management method according to one embodiment disclosed in this document.

[0098] According to one embodiment, a battery management device (e.g., the battery management device (100) of FIG. 1) may perform the operations disclosed in FIG. 5. For example, at least some of the components included in the battery management device (e.g., the memory (110), the control unit (120) of FIG. 1) may be configured to perform the operations of FIG. 5.

[0099] In the following embodiments, the operations of S510 to S530 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation may be changed, and at least two operations may be performed in parallel. Additionally, content corresponding to or overlapping with the above description in relation to FIG. 5 may be briefly explained or omitted.

[0100] Referring to FIG. 5, the battery management method may include the step of obtaining a first charge-discharge profile by performing a first charge-discharge cycle for a first battery cell (S510), the step of inputting input data including the first charge-discharge profile to a 2D CNN (Convolutional Neural Network) trained based on a second charge-discharge profile obtained by performing a second charge-discharge cycle for a second battery cell (S520), and the step of predicting the State of Health (SOH) of the first battery cell based on result data output through the 2D CNN in response to the input data (S530).

[0101] In step S510, the battery management device may charge and discharge the first battery cell a first number of times. For example, the battery management device may control the charging and discharging of the first battery cell based on the first number of times, which was previously defined as the number of target charge-discharge cycles. The first number of times may be, for example, 3 to 10 times, but this is exemplary and the embodiments of the present invention are not limited thereto.

[0102] In step S520, the battery management device can predict the SOH for the first battery cell using a 2D CNN. For example, the 2D CNN may be an artificial intelligence model trained based on second charge-discharge data obtained while charging and discharging a second battery cell, which is distinct from the first battery cell, a second time. The second time may be, for example, 250 to 290 times, but this is exemplary and the embodiments of the present invention are not limited thereto.

[0103] In step S530, the battery management device can predict the SOH of the first battery cell based on result data obtained through a 2D CNN. The battery management device may identify error data including at least one of RMSE, MAE, STD, or any combination thereof based on a comparison result between the prediction result and the actual SOH of the first battery cell, and train the 2D CNN or update the number of target charge / discharge cycles based on the magnitude of the identified error data and whether it exceeds a threshold.

[0105] FIG. 6 is a block diagram showing the hardware configuration of a computing system for performing a method of operation of a battery management device according to one embodiment disclosed in this document.

[0106] Referring to FIG. 6, a computing system (3000) according to one embodiment disclosed in this document may include an MCU (1010), a memory (1020), an input / output I / F (1030), and a communication I / F (1040).

[0107] The MCU (1010) may be a processor that executes various programs stored in memory (1020), processes various information including battery data through these programs, and performs the functions of a processor (or control unit) included in the battery management device shown in FIG. 3 above.

[0108] The memory (1020) can store various programs for performing the functions of the battery management device. Additionally, the memory (1020) can store various information including battery data (voltage data, capacity data, etc.), differential capacity data, etc., and may include a constructed database.

[0109] These memories (1020) may be provided in multiple quantities as needed. The memories (1020) may be volatile memories or non-volatile memories. As volatile memories, the memory (1020) may use RAM, DRAM, SRAM, etc. As non-volatile memories, the memory (1020) may use ROM, PROM, EAROM, EPROM, EEPROM, flash memory, etc. The examples of the memories (1020) listed above are merely examples and are not limited to these examples.

[0110] 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, an output device (not shown), and an MCU (1010).

[0111] The communication I / F (1040) is configured to transmit and receive various data with a server and may be various devices capable of supporting wired or wireless communication. For example, the battery management device may transmit and receive various information, including battery data, from a separately provided external server, etc., through the communication I / F (1040).

[0112] In this way, a computer program according to one embodiment disclosed in this document may be implemented as a module that performs, for example, the functions illustrated in FIG. 1 by being recorded in memory (1020) and processed by an MCU (1010).

[0114] As described above, even though all components constituting the embodiments disclosed in this document have been described as being combined or operating in combination, the embodiments disclosed in this document are not necessarily limited to such embodiments. That is, within the scope of the purposes of the embodiments disclosed in this document, all components may be selectively combined in one or more ways to operate.

[0115] Furthermore, terms such as "include," "compose," or "have" as described above, unless specifically stated otherwise, mean that the relevant component may be inherent; thus, they should be interpreted as allowing for the inclusion of additional components rather than excluding them. All terms, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments disclosed in this document pertain, unless otherwise defined. Commonly used terms, such as those defined in advance, should be interpreted in accordance with their contextual meanings in the relevant technology and, unless explicitly defined in this document, should not be interpreted in an ideal or overly formal sense.

[0116] The above description is merely an illustrative explanation of the technical concept disclosed in this document, and a person skilled in the art to which the embodiments disclosed in this document belong may make various modifications and variations within the scope of the essential characteristics of the embodiments disclosed in this document. Accordingly, the embodiments disclosed in this document are intended to explain, not limit, the technical concept of the embodiments disclosed in this document, and the scope of the technical concept disclosed in this document is not limited by these embodiments. The scope of protection of the technical concept disclosed in this document shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of this document.

Claims

Claim 1 A battery management device comprises: a memory storing at least one instruction; and a control unit operatively connected to the memory; wherein, when the at least one instruction is executed by the control unit, the battery management device: performs a first number of charge-discharge cycles for a first battery cell to obtain first charge-discharge data, and the first charge-discharge data includes one or more data sets corresponding to each of the first number of charge-discharge cycles; A battery management device configured to compare the number of data sets included in the first charge / discharge data with a first number corresponding to the first number of times during a specified period, and if the number of data sets is greater than or equal to the first number corresponding to the first number of times, input the first charge / discharge data into a 2D CNN (Convolutional Neural Network) trained based on second charge / discharge data obtained by performing a second number of charge / discharge cycles for at least one second battery cell, and predict the State of Health (SOH) of the first battery cell at least partially based on the result data output through the 2D CNN corresponding to the first charge / discharge data. Claim 2 A battery management device according to claim 1, characterized in that the first number is less than the second number. Claim 3 A battery management device according to claim 1, wherein the first charge / discharge data comprises one or more data sets relating to voltage data, current data, and temperature data of the first battery cell corresponding to each of the charge / discharge cycles, and each of the one or more data sets corresponds to an instance of a charge / discharge cycle performed in the first battery cell. Claim 4 In claim 3, the battery management device is configured such that, when the at least one instruction is executed by the control unit, the battery management device: determines that at least a portion of the first charge / discharge data is missing or incomplete in response to the determination that the number of the data set is smaller than the first number corresponding to the first number, determines the target number of additional cycle repetitions by calculating the difference between the first number and the number of the data set, and performs additional charge / discharge cycles to obtain third charge / discharge data. Claim 5 In claim 4, the battery management device is configured such that, when the at least one instruction is executed by the control unit, the battery management device inputs the third charge / discharge data into the 2D CNN, generates the result data based at least partially on the third charge / discharge data, and predicts the State of Health (SOH) of the first battery cell based on the result data. Claim 6 A battery management device according to claim 4, characterized in that the target number of additional cycle repetitions is less than or equal to the first number. Claim 7 In claim 1, the battery management device, when the at least one instruction is executed by the control unit, comprises: comparing the actual SOH of the first battery cell with a prediction result generated by the 2D CNN based on the first charge / discharge data, and identifying error data based on the comparison result between the prediction result and the actual SOH, wherein the error data includes at least one of RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), STD (Standard Deviation), or any combination thereof; and using the error data, updating the number of target charge / discharge cycles for SOH prediction or further training the 2D CNN. Claim 8 A battery management method comprising: a control unit performing a first number of charge-discharge cycles for a first battery cell to obtain first charge-discharge data; wherein the first charge-discharge data includes one or more data sets corresponding to each of the first number of charge-discharge cycles; wherein the control unit compares the number of data sets included in the first charge-discharge data with a first number corresponding to the first number during a specified period; wherein, if the number of data sets is greater than or equal to the first number corresponding to the first number, the control unit inputs the first charge-discharge data to a 2D CNN (Convolutional Neural Network) trained based on second charge-discharge data obtained by performing a second number of charge-discharge cycles for at least one second battery cell; and wherein the control unit predicts the State of Health (SOH) of the first battery cell based at least partially on result data output through the 2D CNN corresponding to the first charge-discharge data. Claim 9 A battery management method according to claim 8, characterized in that the first number is less than the second number. Claim 10 In claim 8, the battery management method further comprises: a step in which the control unit determines that at least a portion of the first charge / discharge data is missing or incomplete in response to the determination that the number of the data set is smaller than the first number corresponding to the first number; a step of determining a target number of additional cycle repetitions by calculating the difference between the first number and the number of the data set; and a step of obtaining third charge / discharge data by performing additional charge / discharge cycles in accordance with the target number of additional cycle repetitions. Claim 11 In claim 10, the battery management method further comprises: a step in which the control unit inputs the third charge / discharge data into the 2D CNN; a step of generating the result data based at least partially on the third charge / discharge data; and a step of predicting the State of Health (SOH) of the first battery cell based on the result data. Claim 12 A battery management method according to claim 10, characterized in that the target number of additional cycle repetitions is less than or equal to the first number. Claim 13 In claim 8, the battery management method further comprises: a step in which the control unit compares a prediction result generated by the 2D CNN based on the first charge / discharge data with the actual SOH of the first battery cell; a step of identifying error data based on the comparison result between the prediction result and the actual SOH; the error data includes at least one of RMSE (Root Mean Squared Error), MAE (Mean Absolute Error), STD (Standard Deviation), or any combination thereof; and a step of updating the number of target charge / discharge cycles for SOH prediction or further training the 2D CNN using the error data.

Images