Battery degradation mode prediction system and method, and computer readable recording medium

Abstract

The invention relates to a battery degradation mode prediction system and method, and a computer readable recording medium. The battery degradation mode prediction system includes: a learning data generation unit for generating learning data including a voltage distribution data set for a reference battery and a distribution data set for each degradation mode of the reference battery; the prediction model generation unit is used for generating a prediction model; a data collection unit for obtaining voltage distribution data for the target battery; and a degradation mode analysis unit for obtaining distribution data for each degradation mode of the target battery from the voltage distribution data by using the prediction model, and outputting life data related to a degree of degradation of the target battery for each degradation mode based on the distribution data for each degradation mode. The distribution data for each degradation mode includes cycle-by-cycle degradation trend information for at least one of a degradation parameter associated with the electrode plate and a degradation parameter associated with a resistance of the target battery.

Description

Technical Field

[0001] Various aspects of embodiments of this disclosure relate to battery degradation mode prediction systems and methods. Background Technology

[0002] Unlike primary batteries, which are not designed for (re)charging, secondary (or rechargeable) batteries are designed to discharge and be recharged. Low-capacity secondary batteries are used in portable small electronic devices such as smartphones, feature phones, laptops, digital cameras, and camcorders, while high-capacity secondary batteries are widely used as power sources for driving engines in hybrid and electric vehicles, as well as for storing electricity (e.g., household and / or utility-scale power storage). A secondary battery typically includes an electrode assembly containing positive and negative electrodes, a housing of the electrode assembly, and electrode terminals connected to the electrode assembly.

[0003] Secondary batteries may experience degradation, where their performance declines with increasing usage time. This degradation can occur due to various factors. For example, repeated charge-discharge cycles, excessive current, use at high temperatures, electrolyte decomposition, or deterioration of active materials can be major contributors to accelerated degradation.

[0004] The information disclosed in this background section is intended to enhance the understanding of the background art of this disclosure, and therefore may contain information that does not constitute related (or prior art). Summary of the Invention

[0005] Degradation can lead to reduced capacity, decreased output, or increased internal resistance in rechargeable batteries, and can significantly shorten battery life even after prolonged use. Therefore, accurately identifying and properly managing these degradation modes is desirable in order to maintain the performance and extend the life of rechargeable batteries.

[0006] Embodiments of this disclosure may relate to battery degradation mode prediction systems and methods.

[0007] These and other aspects and features of this disclosure will be described in, or will be apparent from, the following description of embodiments of this disclosure.

[0008] According to one or more embodiments of this disclosure, a battery degradation mode prediction system includes: a learning data generation unit configured to generate learning data, the learning data including a voltage distribution dataset for a plurality of reference batteries and a distribution dataset for each degradation mode of the plurality of reference batteries; a prediction model generation unit configured to generate a prediction model based on the learning data, the prediction model being trained to receive the voltage distribution dataset as input and output a distribution dataset for each degradation mode corresponding to the voltage distribution dataset; a data collection unit configured to obtain voltage distribution data for a target battery; and a degradation mode analysis unit configured to obtain the distribution data for each degradation mode of the target battery from the voltage distribution data using the prediction model, and output lifetime data related to the degree of degradation of the target battery for each degradation mode based on the distribution data for each degradation mode. The distribution data for each degradation mode includes cycle-by-cycle degradation trend information of at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery.

[0009] In an embodiment, the voltage distribution dataset may include full-cell voltage distribution data for multiple reference cells, which is generated by an electrochemical model based on BOL (early life) data and half-cell data associated with the multiple reference cells.

[0010] In an embodiment, half-cell data may include positive half-cell data and negative half-cell data, and full-cell voltage distribution data may be generated based on the resistance characteristics extracted from each of the positive and negative half-cell data.

[0011] In an embodiment, the resistance characteristic may include at least one of material resistance, electrode plate resistance, and diffusion resistance.

[0012] In an embodiment, the data collection unit may be configured to measure the voltage value and charge / discharge capacity of the target battery at predetermined sampling intervals, and obtain voltage distribution data including the correlation between the voltage value and the charge / discharge capacity.

[0013] In an embodiment, the distribution data for each degradation mode may include information evaluated by a prediction model regarding at least one of cycle-by-cycle degradation of positive electrode active material, degradation of negative electrode active material, amount of lithium plating, and increase in side reaction resistance for the target battery.

[0014] In an embodiment, lifetime data may include the degradation rate of the target battery for each degradation mode under a specific cycle.

[0015] In an embodiment, the degradation rate can be expressed as a percentage of the degradation assessment value of the target battery for each degradation mode relative to a reference value.

[0016] In an embodiment, the degradation mode analysis unit can be configured to compare the degradation degree assessment value of the target battery for each degradation mode with a threshold based on the distribution data of the target battery for each degradation mode, and to visually output the degradation degree assessment value based on the comparison result between the degradation degree assessment value and the threshold.

[0017] In an embodiment, the degradation mode analysis unit can be configured to identify the degradation mode whose degradation degree assessment value exceeds a threshold among the multiple degradation modes of the target battery based on the distribution data of the target battery for each degradation mode, and output the main degradation mode.

[0018] In an embodiment, the prediction model may include: an input layer configured to receive voltage distribution data for a target battery as input data; an intermediate layer configured to generate distribution data for each degradation mode of the target battery based on the input data as prediction data; and an output layer configured to output the prediction data.

[0019] In an embodiment, the intermediate layer may include at least one of the following: a first intermediate layer configured to extract features associated with the voltage distribution data; a second intermediate layer configured to process continuous time series characteristics of the voltage distribution data; a third intermediate layer configured to select weights based on the correlation between the voltage distribution data and the distribution data for each degradation mode; and a fourth intermediate layer configured to transform the data dimensions based on information about the number of degradation modes.

[0020] In an embodiment, the second intermediate layer can be configured to receive the output of the first intermediate layer as input to the second intermediate layer.

[0021] According to one or more embodiments of this disclosure, a battery degradation mode prediction method includes: generating learning data, the learning data including a voltage distribution dataset for a plurality of reference batteries and a distribution dataset for each degradation mode of the plurality of reference batteries; generating a prediction model based on the learning data, the prediction model being trained to receive the voltage distribution dataset as input and output a distribution dataset for each degradation mode corresponding to the voltage distribution dataset; obtaining voltage distribution data for a target battery; obtaining the distribution data for each degradation mode of the target battery from the voltage distribution data using the prediction model; and outputting lifetime data related to the degree of degradation of the target battery for each degradation mode based on the distribution data for each degradation mode. The distribution data for each degradation mode includes cycle-by-cycle degradation trend information of at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery.

[0022] In an embodiment, the voltage distribution dataset may include full-cell voltage distribution data for multiple reference cells, which is generated by an electrochemical model based on BOL (early life) data and half-cell data associated with the multiple reference cells.

[0023] In an embodiment, half-cell data may include positive half-cell data and negative half-cell data, and full-cell voltage distribution data may be generated based on the resistance characteristics extracted from each of the positive and negative half-cell data.

[0024] In an embodiment, obtaining voltage distribution data for a target battery may include: measuring the voltage value and charge / discharge capacity of the target battery at predetermined sampling intervals; and obtaining voltage distribution data including the correlation between the voltage value and the charge / discharge capacity.

[0025] In an embodiment, the distribution data for each degradation mode may include information evaluated by a prediction model regarding at least one of cycle-by-cycle degradation of positive electrode active material, degradation of negative electrode active material, amount of lithium plating, and increase in side reaction resistance for the target battery.

[0026] In an embodiment, lifetime data may include the degradation rate of the target battery for each degradation mode under a specific cycle.

[0027] In one embodiment, a non-transitory computer-readable recording medium may store a program that, when executed by a processor, causes the processor to perform the method.

[0028] According to some embodiments of this disclosure, the internal degradation state of a battery can be analyzed or inferred via a battery degradation mode prediction model. Therefore, degradation states can be easily identified without complex processes such as battery disassembly, and degradation trends can be quickly and easily confirmed. Furthermore, the accuracy of battery life prediction can be improved through automated prediction models.

[0029] According to some embodiments of this disclosure, various degradation factors of the battery can be comprehensively considered, thereby enabling more accurate prediction of the battery's internal degradation state. Furthermore, capacity reduction according to each degradation mode can be easily reflected retrospectively, and therefore, after-sales service management of the battery can be performed more effectively.

[0030] However, the aspects and features of this disclosure are not limited to those described above, and other aspects and features not mentioned will be clearly understood by those skilled in the art from the detailed description below. Attached Figure Description

[0031] The accompanying drawings illustrate embodiments of the present disclosure and further describe aspects and features of the disclosure together with the detailed description thereof. Therefore, the present disclosure should not be construed as limited to the drawings.

[0032] Figure 1 The illustration shows an example of a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0033] Figure 2 It is a diagram. Figure 1 Block diagram of a battery degradation mode prediction system.

[0034] Figure 3 This is a block diagram illustrating an information processing system for predicting battery degradation patterns according to an embodiment of the present disclosure.

[0035] Figure 4 This is a diagram illustrating an example of a method for generating learning data in a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0036] Figure 5 This is a diagram illustrating an example of a method for generating a predictive model in a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0037] Figure 6 This is a diagram illustrating an example of a method for predicting a degradation mode for a target battery in a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0038] Figure 7 The illustration shows an example of an artificial neural network model for predicting battery degradation patterns according to embodiments of the present disclosure.

[0039] Figure 8 This is a diagram illustrating an example of a prediction model for a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0040] Figure 9 This is an example of a graph illustrating the distribution data of a target battery for each degradation mode, which is visually output according to an embodiment of the present disclosure.

[0041] Figure 10 This is a diagram illustrating an example of a table in which lifetime data of a target battery, related to the degree of degradation for each degradation mode, is visually shown according to an embodiment of the present disclosure.

[0042] Figure 11 This is a flowchart illustrating a battery degradation mode prediction method according to an embodiment of the present disclosure.

[0043] Description of some figure labels

[0044] 10: Battery Degradation Mode Prediction System

[0045] 120: Voltage distribution data for the target battery

[0046] 140: Lifetime data related to the degree of degradation of the target battery for each degradation mode. Detailed Implementation

[0047] In the following description, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The terms or words used in this specification and claims should not be construed as limited to their general or dictionary meanings, and based on the principle that the inventor can be his / her own lexicographer to appropriately define terms and concepts in order to best interpret his / her invention, they should be interpreted as meanings and concepts consistent with the technical spirit of the present disclosure.

[0048] The embodiments described in this specification and the configurations shown in the accompanying drawings are merely some embodiments of this disclosure and do not represent all technical ideas, aspects, and features of this disclosure. Accordingly, it should be understood that at the time of filing this application, various equivalent schemes and modifications may exist that can replace or modify the embodiments described herein.

[0049] It will be understood that when an element or layer is referred to as being "on," "connected to," or "attached to" another element or layer, it can be directly on, directly connected to, or directly attached to the other element or layer, or one or more intermediary elements or layers may be present. When an element or layer is referred to as being "directly" on, directly connected to, or directly attached to another element or layer, no intermediary element or layer is present. For example, when a first element is described as being "attached" or "connected" to a second element, the first element can be directly attached to or connected to the second element, or the first element can be indirectly attached to or connected to the second element via one or more intermediary elements.

[0050] In the figures, the dimensions of various elements, layers, etc., may be enlarged for clarity. The same reference numerals denote the same elements. As used herein, the term "and / or" includes any and all combinations of one or more of the listed items. Furthermore, when describing embodiments of this disclosure, the use of "may" refers to "one or more embodiments of this disclosure." When preceding / following a list of elements, expressions such as "at least one of" and "any one of" modify the entire list of elements, not individual elements of the list. When phrases such as "at least one of A, B, and C," "at least one selected from the group of A, B, and C," or "at least one selected from A, B, and C" are used to refer to a list of elements A, B, and C, the phrase may refer to any and all suitable combinations or subsets of A, B, and C, such as A, B, C, A and B, A and C, B and C, or A and B and C. As used herein, the terms "use," "being used," and "being exploited" may be considered synonymous with the terms "utilize," "being exploited," and "being exploited," respectively. As used herein, the terms “substantially,” “about,” and similar terms are used as approximations rather than as terms of degree, and are intended to take into account the inherent variations in the measured or calculated values ​​that would be recognized by one of ordinary skill in the art.

[0051] It will be understood that although the terms first, second, third, etc., may be used herein to describe various elements, components, areas, layers, and / or segments, these elements, components, areas, layers, and / or segments should not be limited by these terms. These terms are used to distinguish one element, component, area, layer, or segment from another. Therefore, the first element, component, area, layer, or segment discussed below may be referred to as the second element, component, area, layer, or segment without departing from the teachings of the exemplary embodiments.

[0052] For ease of description, spatial relative terms such as “below,” “under,” “lower,” “above,” and “upper” may be used herein to describe the relationship of one element or feature to another element(s) as illustrated in the figures. It will be understood that, in addition to the orientations depicted in the figures, the spatial relative terms are intended to cover different orientations of the device in use or operation. For example, if the device in the figures is flipped, an element described as “below” or “under” other elements or features will subsequently be oriented “above” or “on” other elements or features. Thus, the term “below” can encompass both above and below orientations. The device may be oriented in other ways (rotated 90 degrees or in other orientations), and the spatial relative descriptors used herein should be interpreted accordingly.

[0053] The terminology used herein is for the purpose of describing embodiments of this disclosure and is not intended to limit this disclosure. As used herein, unless the context clearly indicates otherwise, the singular form "a" and its variations are intended to also include the plural form. It will be further understood that, when used in this specification, the terms "comprising" and / or "including" and their variations specify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0054] Furthermore, any numerical range disclosed and / or described herein is intended to include all subranges with the same numerical precision contained within the described range. For example, the range “1.0 to 10.0” is intended to include all subranges between the described minimum value of 1.0 and the described maximum value of 10.0 (inclusive), that is, having a minimum value greater than or equal to 1.0 and a maximum value less than or equal to 10.0, such as, for example, 2.4 to 7.6. Any maximum numerical limit described herein is intended to include all lower numerical limits contained therein, and any minimum numerical limit described in this specification is intended to include all higher numerical limits contained therein. Accordingly, the applicant reserves the right to amend this specification (including the claims) to expressly describe any subranges contained within the range expressly described herein.

[0055] Referring to two compared elements, features, etc., as “identical” can mean that they are “substantially identical.” Therefore, the phrase “substantially identical” can include cases with deviations considered low in the art, such as less than 5%. Additionally, when a parameter is described as uniform in a given region, this can mean that it is uniform in terms of its mean.

[0056] Throughout the manual, unless otherwise stated, each element may be singular or plural.

[0057] Placing any element "above (or below)" or "on (below)" another element can mean that the arbitrary element can be positioned to contact the upper (or lower) surface of the element, and that the other element can be placed between the element and the arbitrary element positioned on (or below) the element.

[0058] Additionally, it will be understood that when components are referred to as “links,” “connections,” or “attached” to another component, these components can be directly “connected,” “linked,” or “attached” to each other, or another component can be “placed” between these components.

[0059] Throughout the specification, when “A and / or B” is stated, it means A, B, or A and B, unless otherwise stated. That is, “and / or” includes any or all combinations of the listed items. When “C to D” is stated, it means greater than or equal to C and less than or equal to D, unless otherwise specified.

[0060] As used in this article, a reference battery (or secondary battery) may refer to a battery that can generate data for training a degradation pattern prediction model.

[0061] As used in this article, a target battery can refer to a battery for which the degradation mode, degradation factor, degradation rate, or lifetime is predicted by a degradation mode prediction model.

[0062] As used herein, each of the reference battery and the target battery may include at least one battery cell and may be a rechargeable secondary battery. For example, each of the reference battery and the target battery may include a lithium-ion battery.

[0063] Figure 1 The illustration shows an example of a battery degradation mode prediction system 10 according to an embodiment of the present disclosure. Figure 2 It is a diagram. Figure 1 Block diagram of the battery degradation mode prediction system 10.

[0064] In an embodiment, the battery degradation mode prediction system 10 may include a learning data generation unit 210, a prediction model generation unit 220, a data collection unit 230, and a degradation mode analysis unit 240. The battery degradation mode prediction system 10 may receive voltage distribution data 120 for a target battery as input and may output lifetime data 140 related to the degree of degradation of the target battery for each degradation mode (e.g., associated with the degree of degradation of the target battery for each degradation mode).

[0065] In an embodiment, the learning data generation unit 210 can generate learning data for training a battery degradation mode prediction model or a prediction model. The learning data may include a voltage distribution dataset for multiple reference batteries and a distribution dataset for each degradation mode of the multiple reference batteries. Further, the learning data generation unit 210 can generate the learning data using an electrochemical model. This will be discussed in the following references. Figure 4 The learning data generation unit 210, used to generate learning data, is described in more detail.

[0066] In this embodiment, the prediction model generation unit 220 can generate a prediction model. The prediction model generation unit 220 can acquire learning data from the learning data generation unit 210. The prediction model can be trained to identify correlations in the learning data. In other words, the prediction model can be trained to receive a voltage distribution dataset for multiple reference batteries as input and can output a distribution dataset corresponding to the input for each degradation mode. See below for further details. Figure 5 The prediction model generation unit 220 used to generate the prediction model is described in more detail.

[0067] In this embodiment, the data collection unit 230 can obtain measurement data of the target battery, which is the object of the inference. For example, the measurement data of the target battery may include voltage distribution data 120 for the target battery. The data collection unit 230 can directly measure the target battery to obtain real-time data, or it can obtain predicted data from a battery management system associated with the target battery. Reference will be made below. Figure 6 The data collection unit 230 for obtaining measurement data of the target battery is described in more detail.

[0068] In an embodiment, the degradation mode analysis unit 240 can obtain distribution data for each degradation mode of the target battery. The degradation mode analysis unit 240 can access the prediction model generated by the prediction model generation unit 220. Additionally, the degradation mode analysis unit 240 can obtain measurement data of the target battery from the data collection unit 230. Therefore, the degradation mode analysis unit 240 can use the prediction model to obtain distribution data for each degradation mode of the target battery from the measurement data of the target battery. The distribution data for each degradation mode may include cycle-by-cycle degradation trend information for at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery. Reference will be made below. Figure 6 The degradation mode analysis unit 240, used to obtain distribution data for each degradation mode of the target battery, is described in more detail.

[0069] In an embodiment, the degradation mode analysis unit 240 can output lifetime data 140 related to the degree of degradation of the target battery for each degradation mode. The degradation mode analysis unit 240 can output lifetime data related to the degree of degradation of the target battery for each degradation mode based on the distribution data of the target battery for each degradation mode obtained using a prediction model. This will be referenced below. Figure 6 Provide a more detailed description of the lifespan data.

[0070] With the above configuration, the battery degradation mode prediction system 10 can analyze or infer the internal degradation state of the battery using a battery degradation mode prediction model. Accordingly, degradation states can be identified more easily without complex processes such as battery disassembly, and degradation trends can be quickly and easily confirmed. Furthermore, the accuracy of battery life prediction can be improved through an automated prediction model.

[0071] Furthermore, the above configuration allows for a comprehensive consideration of various battery degradation factors, enabling more accurate prediction of the battery's internal degradation state. Additionally, it makes it easier to reflect capacity reduction after each degradation mode, facilitating more effective after-sales battery management.

[0072] Figure 3 This is a block diagram illustrating an information processing system 300 for predicting battery degradation patterns according to an embodiment of the present disclosure.

[0073] Information processing system 300 can be with Figure 1 and Figure 2 This corresponds to the battery degradation mode prediction system 10 illustrated in the diagram. The information processing system 300 may include a memory 310, a processor 320, a communication module 330, and an input / output interface 340. The information processing system 300 can transmit information and / or data via a network using the communication module 330. The information processing system 300 may consist of at least one device including the memory 310, processor 320, communication module 330, and input / output interface 340.

[0074] The memory 310 may include any suitable non-transitory computer-readable recording medium. According to embodiments, the memory 310 may include a persistent mass storage device such as ROM (Read-Only Memory), a disk drive, an SSD (Solid State Drive), or flash memory. As another example, a persistent mass storage device such as ROM, SSD, flash memory, or disk drive may be included in the information processing system 300 as a separate persistent storage device distinct from the memory 310. Additionally, the memory 310 may store software components including an operating system and at least one program code (e.g., code for generating and training a degradation mode prediction model installed and driven in the information processing system 300, and code for predicting the degradation mode of a target battery using the prediction model).

[0075] Software components can be loaded from a separate computer-readable recording medium different from memory 310. This separate computer-readable recording medium can be a recording medium that can be directly connected to information processing system 300, such as a floppy disk drive, magnetic disk, magnetic tape, DVD / CD-ROM drive, or memory card. As another example, software components can be loaded into memory 310 via communication module 330, which may not be a computer-readable recording medium. For example, at least one program can be loaded into memory 310 based on a file provided by a developer or by a file distribution system distributing installation files of an application via communication module 330. This program may include, for example, programs for generating and training degradation mode prediction models and programs for predicting degradation modes of a target battery using the prediction models.

[0076] Processor 320 can process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. Instructions can be provided from a user terminal or another external system by memory 310 or communication module 330. For example, processor 320 can receive learning data from one or more external devices or battery manufacturing facilities, including a voltage distribution dataset for a reference battery, a distribution dataset for each degradation mode, and / or voltage distribution data for a target battery, as well as data for various battery types. Processor 320 can generate and train a degradation mode prediction model based on the learning data or based on the voltage distribution data of the target battery. The prediction model can be used to calculate the distribution data for each degradation mode of the target battery and / or lifetime data related to the degree of degradation for each degradation mode.

[0077] The communication module 330 can provide configuration or functions to enable the user terminal and the information processing system 300 to communicate with each other via a network, and can also provide configuration or functions to enable the information processing system 300 to communicate with external systems (e.g., independent battery manufacturing process facilities and cloud systems). For example, under the control of the processor 320 of the information processing system 300, control signals, commands, and data provided can be transmitted to the user terminal and / or external system via the communication module 330 and the network through the communication module 330 and the communication module of the user terminal and / or the external system. For example, the life status of the target battery generated by the information processing system 300 can be transmitted to the user terminal and / or the external system via the communication module 330 and the network through the communication module 330 and the communication module of the user terminal and / or the external system. In addition, the user terminal and / or external system receiving the life status of the target battery can output the received information through a device capable of displaying the output.

[0078] The input / output interface 340 of the information processing system 300 can be a device for interacting with a device for input or output, which can be connected to the information processing system 300 or included in the information processing system 300. Figure 3 In the illustration, the input / output interface 340 is shown as a component separate from the processor 320, but this disclosure is not limited thereto. The input / output interface 340 may be included within the processor 320. The information processing system 300 may include more than Figure 3 The component shown is the one with more parts.

[0079] The processor 320 of the information processing system 300 can manage, process, and / or store information and / or data received from multiple user terminals and / or multiple external systems. According to an embodiment, the processor 320 can receive learning data associated with a reference battery and / or voltage distribution data of a target battery, as well as lifetime data for various types of batteries, from user terminals and / or external systems. The processor 320 can use a degradation mode prediction model to calculate distribution data and / or lifetime data for each degradation mode of the target battery based on the voltage distribution data of the target battery, and can output the calculated lifetime data, etc., through a device connected to the information processing system 300 capable of displaying the output.

[0080] Figure 4 This is a diagram illustrating an example of a method 400 for generating learning data 420 in a battery degradation mode prediction system according to an embodiment of the present disclosure. In the following, reference is made to... Figure 4 It is not necessary to repeat the above reference. Figures 1 to 3 The description is redundant.

[0081] In this embodiment, the learning data generation unit 210 can generate learning data 420. Learning data 420 may include a voltage distribution dataset 422 for multiple reference batteries and / or a distribution dataset 424 for each degradation mode of multiple reference batteries. Learning data 420 can be provided to the prediction model generation unit 220 as base data for training a battery degradation mode prediction model.

[0082] In an embodiment, the voltage distribution dataset 422 may include voltage and capacity data, such as those measured during multiple lifetime cycles (e.g., a predetermined number of lifetime cycles) for each of a plurality of reference cells. A reference cell may be a full cell in which data is collected over multiple lifetime cycles from BOL (beginning of life) to EOL (end of life). For example, data may be measured in increments of 100 cycles over 1000 lifetime cycles.

[0083] In an embodiment, the distribution dataset 424 for each degradation mode may include cycle-by-cycle voltage and capacity variation information (e.g., predetermined cycle-by-cycle voltage and capacity variation information) based on the degradation mode (or degradation factor). A degradation mode may refer to a physicochemical change or degradation mechanism of the internal configuration that affects battery performance. For example, it may include the positive electrode active material, the negative electrode active material, the amount of lithium plating on the negative electrode, and the increase in side reaction resistance.

[0084] In an embodiment, the learning data generation unit 210 can generate learning data 420 based on a BOL dataset 412 and a half-cell dataset 414 associated with multiple reference cells. The BOL dataset 412 can include electrochemical characteristics (such as voltage, charge capacity, or internal resistance) of the reference cells (which are full cells) in their initial manufacturing state. This allows the initial performance of the reference cells to be used as a baseline for tracking internal degradation processes, thus forming the basis data for predictive models. The half-cell dataset 414 can include positive electrode half-cell data and negative electrode half-cell data, as well as independent analytical data for each of the positive and negative electrodes. Therefore, variations in the materials, specifications, arrangement, and electrode plates and active materials of each electrode can be reflected, clearly identifying electrochemical changes and performance degradation that may occur in each electrode. Because the BOL dataset 412 and the half-cell dataset 414 for full cells are considered together when generating the learning data, battery degradation mechanisms can be analyzed and predicted more accurately.

[0085] In this embodiment, the learning data generation unit 210 can generate learning data 420 through an electrochemical model. For example, various resistance characteristics of the half-cell can be extracted separately using the electrochemical model, and the full-cell voltage distribution can be generated based on these resistance characteristics. Accordingly, the electrochemical interactions within the battery can be modeled, and the battery degradation process can be analyzed and predicted more accurately.

[0086] In an embodiment, the resistance characteristics may include at least one of the material resistance, electrode plate resistance, and diffusion resistance extracted from each of the positive and negative electrode half-cells. Full-cell voltage distribution data for multiple reference cells can be generated based on the extracted resistance characteristics. The material resistance may include resistance related to the conductivity of the positive and negative electrode active materials. The electrode plate resistance may be associated with electrochemical reactions occurring on the electrode surface and can be the basis for analyzing charge-discharge efficiency. The diffusion resistance may include resistance associated with the migration rate of lithium ions.

[0087] Figure 5 This is a diagram illustrating an example of a method 500 for generating a prediction model 510 in a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0088] In an embodiment, the prediction model generation unit 220 can train and generate a prediction model 510 such that, upon receiving a voltage distribution dataset 522 for multiple reference batteries as input, it outputs a distribution dataset 524 corresponding to the voltage distribution dataset 522 for each degradation mode. The voltage distribution dataset 522 and the distribution dataset 524 for each degradation mode can be coupled with... Figure 4 The learning data 420 generated by the learning data generation unit 210 includes a voltage distribution dataset 422 and a distribution dataset 424 for each degradation mode. In the following text, reference is made to... Figure 5 It is not necessary to repeat the above reference. Figures 1 to 4 The description is redundant.

[0089] In an embodiment, the prediction model generation unit 220 can train the correlation between a voltage distribution dataset 522 during multiple lifetime cycles (e.g., a predetermined number of lifetime cycles) of a reference battery and a corresponding distribution dataset 524 for each degradation mode. The prediction model 510 can learn the degradation trend of the reference battery through each degradation mode in order to predict the internal degradation state of the target battery.

[0090] In an embodiment, the prediction model 510 may include an LSTM (Long Short-Term Memory)-based machine learning model that can optimize or improve the processing of time-series data. Therefore, continuous voltage variation patterns included in the voltage distribution data and their correlation with the distribution data for each degradation pattern can be analyzed more effectively. Furthermore, optimization algorithms (e.g., Bayesian optimizers) can be used to optimize the model's parameters and structure, thereby maximizing or improving the performance of the prediction model 510. However, the prediction model 510 is not limited to LSTM, and references will be made below to other models. Figure 7 and Figure 8 Provide a more detailed description of prediction model 510.

[0091] Figure 6 This is a diagram illustrating an example of a method 600 for predicting a degradation mode for a target battery in a battery degradation mode prediction system according to an embodiment of the present disclosure. In the following, reference is made to... Figure 6 It is not necessary to repeat the above reference. Figures 1 to 5 The description is redundant.

[0092] In an embodiment, the data collection unit 230 can obtain voltage distribution data 610 for the target battery. The data collection unit 230 can directly measure the voltage value and charge / discharge capacity of the target battery at appropriate sampling intervals (e.g., at predetermined sampling intervals) to obtain voltage distribution data 610 including the correlation between the measured voltage value and the charge / discharge capacity. However, the method for obtaining the voltage distribution data 610 is not limited to this. For example, the data collection unit 230 can obtain the voltage distribution data 610 via a communication module (e.g., Figure 3 The data collection unit 230 collects voltage and charge / discharge data of the target battery from a battery management system associated with the target battery or an external measurement device. Alternatively, the data collection unit 230 can obtain data about the target battery through a cloud-based measurement system.

[0093] In an embodiment, the degradation mode analysis unit 240 can receive voltage distribution data 610 of the target battery as input and use the prediction model 510 to predict distribution data 620 for each degradation mode of the target battery. The degradation mode can include major factors affecting the degradation of the target battery (such as various degradation parameters associated with the battery's electrode plates or resistance). For example, the degradation mode can include the amount of degradation of the positive electrode active material, the amount of degradation of the negative electrode active material, the amount of lithium plating, and the increase in resistance due to side reactions. Additionally, the distribution data 620 for each degradation mode can include information reflecting the cycle-by-cycle degradation trend of the target battery according to the degradation mode. For example, it can include data showing the rate of degradation of the positive and negative electrode active materials for each cycle of battery use, capacity reduction due to lithium plating, or increased internal resistance due to side reactions over time. (Refer to below) Figure 9 A more detailed example of visually outputting the distribution data for each degradation pattern 620 is provided.

[0094] In an embodiment, the degradation mode analysis unit 240 may, based on voltage distribution data 610 and / or distribution data 620 for each degradation mode, use a prediction model 510 to output lifetime data 630 related to the degree of degradation of the target battery for each degradation mode. The lifetime data 630 may include information for quantitatively assessing the degradation state of the target battery and predicting its remaining lifetime. See below for reference. Figure 10 A more detailed example of visually outputting lifetime data 630 is provided.

[0095] In an embodiment, degradation reference values ​​and thresholds for the target battery can be defined in relation to the degree of degradation or lifetime data of the target battery according to the degradation mode. The degradation thresholds and / or reference values ​​for the target battery for each degradation mode can be derived using learning data based on multiple reference batteries during the training of the prediction model 510. As another example, the degradation reference values ​​and thresholds can be determined (e.g., predetermined) or obtained by user input.

[0096] The reference value can be a numerical value within the normal operating range of the battery, representing the permissible level of degradation at which the battery can operate safely and stably under normal degradation conditions. Degradation within the reference value range can be considered not to significantly affect battery performance, and if the assessed degradation level of the target battery is included within such a range (e.g., when the assessed degradation level of the target battery is included within such a range), it can be predicted that the battery will still operate stably.

[0097] On the other hand, a threshold can be defined as the limit that may severely affect battery performance or lifespan. Under a given degradation mode, if the degree of degradation exceeds the threshold (e.g., when the degree of degradation exceeds the threshold), this could mean that battery performance begins to deteriorate rapidly, or that there is a possibility of safety issues. For example, if it is determined that the degradation of the positive electrode active material assessed by a predictive model exceeds the threshold (e.g., when it is determined that the degradation of the positive electrode active material assessed by a predictive model exceeds the threshold), it can be predicted that immediate after-sales maintenance management of the positive electrode active material may be required for the performance or safety of the target battery.

[0098] In an embodiment, lifetime data 630 may include the degradation rate of the target battery for each degradation mode, evaluated by prediction model 510 under specific cycles. The degradation rate may include a value obtained by expressing the degree of degradation evaluated in each degradation mode as a percentage compared to a reference value used for battery performance. The degradation rate allows for a more accurate assessment of the degradation process of the target battery.

[0099] In an embodiment, the degradation mode analysis unit 240 can compare the degradation severity assessment value of the target battery for each degradation mode with a threshold based on the distribution data 620 of the target battery for each degradation mode, and can visually output the degradation severity assessment value according to the comparison result between the degradation severity assessment value and the threshold. For example, if it is determined that the degradation severity assessment value for a certain degradation mode of the target battery is close to or exceeds the threshold (e.g., when it is determined that the degradation severity assessment value for a certain degradation mode of the target battery is close to or exceeds the threshold), the degradation severity assessment value can be output in a visually emphasized manner. A visualization dashboard or graph can intuitively convey the assessed degradation status of the target battery to the user, and degradation modes that exceed the threshold can be displayed with a warning color (e.g., red) to indicate that immediate management may be required.

[0100] In an embodiment, the degradation pattern analysis unit 240 can identify degradation patterns whose degradation severity assessment values ​​exceed a threshold as primary degradation patterns based on distribution data 620 for each degradation pattern, and can output the primary degradation patterns. For example, the primary degradation patterns can be displayed in order of the largest number of degradation severity assessment values ​​exceeding the threshold. In such cases, the primary degradation patterns can be displayed in a distribution graph or table, which can be differentiated by color depending on the degree to which the threshold is exceeded. For example, in a graph or table, degradation patterns whose degradation severity estimates exceed the threshold by more than an appropriate numerical range (e.g., a predetermined numerical range) can be displayed in red, medium ranges in yellow, and small ranges in green.

[0101] Therefore, users can intuitively grasp the severity of the degradation state of the target battery for each degradation mode, and can more easily identify degradation modes that require urgent management.

[0102] Figure 7 The illustration shows an example of an artificial neural network model 700 for predicting battery degradation patterns according to an embodiment of the present disclosure.

[0103] Artificial neural network model 700 can refer to a statistical learning algorithm or the execution structure of an algorithm implemented based on machine learning technology and biological neural network structures in cognitive science. Artificial neural network model 700 can be compared with... Figure 5 The prediction model 510 corresponds to this.

[0104] In an embodiment, the artificial neural network model 700 may be a machine learning model in which nodes (e.g., artificial neurons) forming a network via synaptic connections repeatedly adjust synaptic weights such that the error between the correct output and the inferred output corresponding to a specific input can be reduced, thereby possessing problem-solving capabilities. For example, the artificial neural network model 700 may include any suitable stochastic model and / or neural network model used in artificial intelligence learning methods such as machine learning or deep learning.

[0105] In an embodiment, the artificial neural network model 700 can be implemented as a multilayer perceptron (MLP) consisting of multiple node layers and connections between them. For example, the artificial neural network model 700 can be implemented using one of a variety of suitable artificial neural network model structures that include an MLP.

[0106] like Figure 7As shown, the artificial neural network model 700 can consist of an input layer 720, an output layer 740, and n hidden layers 730_1 to 730_n (where n is a natural number). The input layer 720 receives input data including voltage distribution data 710 for the target battery. The output layer 740 outputs distribution data 750 of the target battery for each degradation mode as output data corresponding to the input data. The n hidden layers 730_1 to 730_n are located between the input layer 720 and the output layer 740, receiving signals from the input layer 720, extracting features, and transmitting the features to the output layer 740. The output layer 740 can receive signals from the hidden layers 730_1 to 730_n and can output signals to the outside.

[0107] In embodiments, the learning method of the artificial neural network model 700 may include a supervised learning method and an unsupervised learning method that does not use or requires a teacher signal. In the supervised learning method, the model is trained to optimize problem solving by inputting a teacher signal (e.g., a correct answer). As an example, the information processing system may use a voltage distribution dataset for multiple reference batteries and a distribution dataset for each degradation mode of the multiple reference batteries as learning data. Supervised learning may be performed to analyze multiple reference batteries using the correspondence between the voltage distribution dataset and the distribution dataset for each degradation mode as ground truth data. The artificial neural network model 700 may be trained to infer the distribution data for each degradation mode of the target battery corresponding to the voltage distribution data of the target battery and / or the lifetime data related to the degree of degradation of the target battery for each degradation mode.

[0108] In an embodiment, the information processing system can train an artificial neural network model 700 such that the difference between the distribution dataset of the reference battery predicted by the artificial neural network model 700 for each degradation mode and the true value data is minimized or reduced.

[0109] Therefore, multiple input variables and their corresponding multiple output variables can be matched to the input layer 720 and output layer 740 of the artificial neural network model 700, respectively. By adjusting the synaptic values ​​in the nodes of the input layer 720, hidden layers 730_1 to 730_n, and output layer 740, the correct output corresponding to a specific input can be extracted. Through the learning process, the artificial neural network model 700 can identify the characteristics hidden in the input variables and adjust the synaptic values ​​(e.g., weights) in the nodes of the artificial neural network model 700 to reduce the error between the output variable calculated based on the input variables and the target output.

[0110] With this configuration, the battery degradation mode prediction system 10 can appropriately evaluate the distribution data of the target battery for each degradation mode and / or the lifetime data related to the degree of degradation for each degradation mode by using an artificial neural network model 700 trained on multiple reference batteries and input data 710 based on the distribution data from the target battery.

[0111] Figure 8 This is a diagram illustrating an example of a prediction model 800 of a battery degradation mode prediction system according to an embodiment of the present disclosure.

[0112] The prediction model 800 may include an input layer 810, an intermediate layer 820, and an output layer 830, and can be coupled with... Figure 7 This corresponds to the artificial neural network model 700. See below for reference. Figure 8 It is not necessary to repeat the above reference. Figures 1 to 7 The description is redundant.

[0113] In one embodiment, the input layer 810 may receive voltage distribution data for the target battery. The input data may include time-series correlation information between the voltage variations of the target battery and its charge / discharge capacity. Therefore, basic data for predicting the internal degradation state of the battery can be provided.

[0114] In one embodiment, the output layer 830 can output predicted data generated based on the input data. The predicted data may include distribution data for each degradation mode of the target battery. In another embodiment, the predicted data may further include lifetime data based on the degree of degradation of the target battery for each degradation mode. Therefore, information about the battery's internal degradation state and remaining lifetime can be effectively provided.

[0115] In an embodiment, intermediate layer 820 can generate predictive data for the target battery based on the input data. Intermediate layer 820 may include one or more detailed intermediate layers. Each of the detailed intermediate layers can perform a specific role. For example, to analyze the time-series characteristics of the input data, LSTM (Long Short-Term Memory) or Bi-LSTM (Bidirectional LSTM) can be used as a second intermediate layer 824. Therefore, the continuous characteristics of the voltage distribution data can be analyzed effectively.

[0116] In an embodiment, intermediate layer 820 may include a first intermediate layer 822 that allows features and patterns of the input data to be extracted. For example, the process may include extracting information such as the rate of change of the target battery's voltage value, variability patterns, outliers, asymmetry, and peak values. To extract these features, the first intermediate layer 822 may include a CNN (Convolutional Neural Network) model that performs convolution operations on the input data.

[0117] In this embodiment, the first intermediate layer 822 may be located upstream of the second intermediate layer 824. Furthermore, the output of the first intermediate layer 822 may be transmitted as input to the second intermediate layer 824. Therefore, predictions reflecting the characteristics and time-series properties of the voltage distribution data are possible, and the performance and accuracy of the prediction model can be improved.

[0118] In an embodiment, intermediate layer 820 may include an attention mechanism as a third intermediate layer 826. The third intermediate layer 826 may analyze the correlation between input and output data and assign high weights to information with high cross-correlation. For example, during the learning process, the correlation between voltage distribution datasets of multiple reference batteries and distribution datasets for each degradation mode may be analyzed, and activations associated with voltage distribution data that are highly correlated with the output value may be assigned high weights. Accordingly, the accuracy of predicting battery degradation modes or internal degradation trends of the battery can be improved.

[0119] In an embodiment, intermediate layer 820 may include a fourth intermediate layer (e.g., a flattening layer) 828 for transforming the dimensions of the output data. The fourth intermediate layer 828 may transform the dimensions of the output data based on the number of degradation modes. For example, if the internal degradation state of the battery is analyzed using four degradation modes (such as the degree of degradation of the positive electrode active material and the degree of degradation of the negative electrode active material, the amount of lithium plating, and the increase in resistance of the side reaction) (e.g., when the internal degradation state of the battery is analyzed using four degradation modes (such as the degree of degradation of the positive electrode active material and the degree of degradation of the negative electrode active material, the amount of lithium plating, and the increase in resistance of the side reaction), then the fourth intermediate layer 828 can transform the output data according to the number of degradation modes (e.g., four). Therefore, more accurate analysis results can be derived efficiently.

[0120] The above structure provides a multi-level intermediate layer structure in which the battery degradation prediction model can comprehensively consider the complex characteristics of time series data and various degradation factors, thereby improving the accuracy and efficiency of predicting the internal degradation state of the battery.

[0121] Figure 9 These are examples of graphs 900 and 902 illustrating, according to embodiments of the present disclosure, in which distribution data of a target battery for each degradation mode are visually output as graphs 900 and 902.

[0122] Figure 9 The curves 900 and 902 can be correlated with the distribution data 620 of the target battery for each degradation mode, which can be derived from... Figure 6 The degradation mode analysis unit 240 in the model is calculated using the prediction model 510. (Refer to the following text.) Figure 9 It is not necessary to repeat the above reference. Figures 1 to 8The description is redundant.

[0123] The first curve (Figure 900) visually illustrates how the charging capacity of the negative electrode active material changes relative to a reference value at each cycle. Therefore, the degradation trend of the negative electrode active material can be easily identified. Furthermore, depending on the progress of battery degradation over time, the impact of negative electrode active material degradation on battery performance can be clearly analyzed.

[0124] The second graph, 902, numerically illustrates the increasing pattern of lithium plating amount assessed with each cycle. Therefore, it visually confirms how the lithium plating amount increases over time and allows for precise analysis of the impact of lithium plating on battery degradation.

[0125] As mentioned above, the degradation trend of the battery according to each degradation mode can be easily identified, and the battery degradation status can be monitored in real time or the main causes of battery performance degradation can be accurately analyzed.

[0126] Figure 10 This is a diagram illustrating an example of a table 1000 in which lifetime data related to the degree of degradation of a target battery for each degradation mode is visually shown according to an embodiment of the present disclosure.

[0127] Figure 10 Table 1000 may correspond to lifetime data 620 related to the degree of degradation of the target battery for each degradation mode, and lifetime data 620 may be used by degradation mode analysis unit 240. Figure 6 The prediction model 510 is used for calculation. See below for reference. Figure 10 It is not necessary to repeat the above reference. Figures 1 to 9 The description is redundant.

[0128] Table 1000 visually illustrates the main degradation modes of the target battery and their degradation rates. Each degradation mode may include the amount of degradation of the positive electrode active material, the amount of degradation of the negative electrode active material, the amount of lithium plating on the negative electrode, and the increase in side reaction resistance, etc., and the degradation rate for each degradation mode can provide lifetime data related to the performance degradation of the target battery.

[0129] In an embodiment, multiple degradation modes may be displayed in order of higher degradation rates. In an embodiment, if the degradation rate exceeds a threshold (e.g., a specific or predetermined threshold) (e.g., when the degradation rate exceeds a threshold (e.g., a specific or predetermined threshold)), it may be displayed in a visually emphasized manner.

[0130] Accordingly, the battery degradation process according to each degradation mode can be easily identified, thereby enabling effective battery management and performance prediction.

[0131] Figure 11This is a flowchart illustrating a battery degradation mode prediction method 1100 according to an embodiment of the present disclosure.

[0132] In an embodiment, the battery degradation mode prediction method 1100 may be executed by at least one processor of a user terminal and / or an information processing system. Hereinafter, references are made to... Figure 11 It is not necessary to repeat the above reference. Figures 1 to 10 The description is redundant.

[0133] In an embodiment, the battery degradation mode prediction method 1100 can begin when the processor generates learning data, which includes a voltage distribution dataset for multiple reference batteries and a distribution dataset for each degradation mode of the multiple reference batteries (S1110). The voltage distribution dataset for the multiple reference batteries may include full-cell voltage distribution data for the multiple reference batteries, which can be generated by an electrochemical model based on BOL (early life) data and half-cell data associated with the multiple reference batteries. Additionally, the half-cell data may include positive electrode half-cell data and negative electrode half-cell data, and the full-cell voltage distribution data may be generated based on resistive characteristics extracted from each of the positive electrode half-cell data and the negative electrode half-cell data.

[0134] The processor can generate a prediction model based on the learning data. The prediction model is trained to receive a voltage distribution dataset as input and outputs a distribution dataset corresponding to the voltage distribution dataset for each degradation mode (S1120).

[0135] Subsequently, the processor can obtain voltage distribution data for the target battery (S1130). The processor can measure the voltage value and charge / discharge capacity of the target battery at an appropriate sampling interval (e.g., a predetermined sampling interval), and can obtain voltage distribution data including the correlation between the measured voltage value and the charge / discharge capacity.

[0136] The processor can obtain distribution data for each degradation mode of the target battery from the voltage distribution data using a prediction model (S1140). The distribution data for each degradation mode may include cycle-by-cycle degradation trend information for at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery. For example, the distribution data for each degradation mode may include information that can be evaluated by the prediction model regarding at least one of cycle-by-cycle degradation of the positive electrode active material, degradation of the negative electrode active material, lithium plating amount, and increase in side reaction resistance for the target battery.

[0137] The processor can then output lifetime data related to the degree of degradation of the target battery for each degradation mode, based on the distribution data for each degradation mode (S1150). The lifetime data may include the degradation rate of the target battery for each degradation mode under a specific cycle.

[0138] The methods described above can be provided as computer programs stored in a computer-readable recording medium for execution in a computer. The medium may continuously store the computer-executable program or temporarily store it for execution or download. The medium can be various recording or storage devices in the form of a single or multiple hardware components, and is not limited to media directly connected to a particular computer system. The medium may also be distributed across a network. Examples of media include magnetic media (such as hard disks, floppy disks, or magnetic tapes), optical recording media (such as CD-ROMs or DVDs), magneto-optical media (such as optical floppy disks), and ROM, RAM, and / or flash memory, which can store program instructions. Other examples of media may include storage or recording media managed by application stores that distribute applications or by various other software provisioning or distribution sites or servers.

[0139] The methods, operations, techniques, and apparatuses of this disclosure (e.g., learning data generation units, predictive model generation units, data collection units, and degradation pattern analysis units, etc.) can be implemented by various suitable means. For example, they can be implemented via hardware, firmware, software, or a combination thereof. Those skilled in the art will understand that the various logic blocks, modules, circuits, and algorithmic processes described in connection with this disclosure can be implemented in electronic hardware, computer software, or a combination of both. To more clearly illustrate the interchangeability of hardware and software, the various components, blocks, modules, circuits, and processes have been generally described above from a functional perspective. Whether such functionality is implemented in hardware or software depends on the specific application and the design requirements of the overall system. Those skilled in the art can implement the described functions in various ways for each specific application, but such implementations should not be construed as limiting the embodiments according to this disclosure.

[0140] In a hardware implementation, the processing unit for performing these techniques may be implemented in one or more ASICs, DSPs, digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic devices, or other electronic units, computers, or combinations thereof designed to perform the functions described in this disclosure.

[0141] Accordingly, the various logic blocks, modules, and circuits described in this disclosure can be implemented or executed using a general-purpose processor, DSP, ASIC, FPGA, or other suitable programmable logic device, discrete gate or transistor logic, discrete hardware components, or any suitable combination thereof, designed to perform the functions described in this disclosure. The general-purpose processor may be a microprocessor, but alternatively, as those skilled in the art will understand, the processor may be any suitable processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other suitable configuration.

[0142] In firmware and / or software implementations, these techniques can be implemented as instructions stored on a computer-readable medium such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, optical disc (CD), or magnetic or optical data storage devices. The instructions can be executed by one or more processors, causing the processors to perform certain aspects of the functions described herein.

[0143] When implemented in software, these technologies may be stored as one or more instructions or code on or transmitted via a computer-readable medium. Computer-readable media includes both storage and communication media, encompassing any suitable medium that facilitates the transfer of a computer program from one location to another, and covering any suitable and usable medium accessible to a computer. Storage media may be any suitable and usable medium accessible to a computer. Non-limiting examples include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage, or any other suitable medium that can be used to carry or store the required program code in the form of instructions or data structures, and which is accessible to a computer. Any connection is also appropriately referred to as a computer-readable medium.

[0144] For example, when software is transmitted from a website, server, or other remote source via coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, these technologies are all included within the definition of media. As used herein, the terms disk and disc include compact disc (CD), laser disc, optical disc, DVD (Digital Versatile Disc), floppy disk, and Blu-ray disc, wherein disks typically magnetically copy data, while discs optically copy data using lasers. Various combinations of the above can also be included as computer-readable media.

[0145] Software modules can reside in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disks, removable disks, CD-ROMs, or any other suitable form of storage medium. The storage medium can be coupled to the processor, allowing the processor to read information from or write information to the storage medium. Alternatively, the storage medium can be integrated into the processor. The processor and storage medium can reside in an ASIC. This ASIC can reside in the user terminal. Alternatively, the processor and storage medium can exist as separate components in the user terminal.

[0146] While some embodiments have been described as utilizing one or more independent computer systems, this disclosure is not limited thereto and can be implemented in conjunction with any suitable computing environment, such as a networked or distributed computing environment. Furthermore, some embodiments of this disclosure can be implemented in multiple processing chips or devices, and storage can be similarly affected across multiple devices. These devices may include PCs, web servers, and portable devices.

[0147] While some embodiments of this disclosure have been described, various suitable modifications and alterations can be made by those skilled in the art without departing from the spirit and scope of this disclosure. It should also be understood that such modifications and alterations fall within the spirit and scope of the claims.

[0148] For illustrative purposes, the embodiments of the present disclosure described above have been disclosed, and those skilled in the art will understand that various modifications, alterations, and additions are possible within the spirit and scope of the present disclosure, and such modifications, alterations, and additions should be considered to fall within the scope of the claims.

[0149] Those skilled in the art to which this disclosure pertains will understand that various substitutions, modifications, and alterations are possible without departing from the spirit and scope of this disclosure. Therefore, this disclosure is not limited to the above embodiments and the accompanying drawings.

[0150] Although this disclosure has been described with reference to the accompanying drawings illustrating various aspects of the embodiments, this disclosure is not limited thereto. Those skilled in the art to which this disclosure pertains can make various modifications and alterations within the spirit of this disclosure and within the scope of the following claims and their equivalents.

Claims

1. A battery degradation mode prediction system, comprising: A learning data generation unit is configured to generate learning data, which includes a voltage distribution dataset for multiple reference cells and a distribution dataset for each degradation mode of the multiple reference cells. A prediction model generation unit is configured to generate a prediction model based on the learning data, the prediction model being trained to receive the voltage distribution dataset as input and output the distribution dataset corresponding to the voltage distribution dataset for each degradation mode; The data collection unit is configured to acquire voltage distribution data for the target battery; as well as The degradation mode analysis unit is configured to obtain distribution data for each degradation mode of the target battery from the voltage distribution data using the prediction model, and to output lifetime data related to the degree of degradation of the target battery for each degradation mode based on the distribution data for each degradation mode. The distribution data for each degradation mode includes cycle-by-cycle degradation trend information for at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery.

2. The battery degradation mode prediction system according to claim 1, wherein, The voltage distribution dataset includes full-cell voltage distribution data for the plurality of reference cells, which is generated by an electrochemical model based on early lifetime data and half-cell data associated with the plurality of reference cells.

3. The battery degradation mode prediction system according to claim 2, wherein, The half-cell data includes positive electrode half-cell data and negative electrode half-cell data, and The full-cell voltage distribution data is generated based on the resistance characteristics extracted from each of the positive half-cell data and the negative half-cell data.

4. The battery degradation mode prediction system according to claim 3, wherein, The resistance characteristics include at least one of material resistance, electrode plate resistance, and diffusion resistance.

5. The battery degradation mode prediction system according to claim 1, wherein, The data collection unit is configured to measure the voltage value and charge / discharge capacity of the target battery at predetermined sampling intervals, and to obtain voltage distribution data including the correlation between the voltage value and the charge / discharge capacity.

6. The battery degradation mode prediction system according to claim 1, wherein, The distribution data for each degradation mode includes information evaluated by the prediction model regarding at least one of cycle-by-cycle degradation of the positive electrode active material, degradation of the negative electrode active material, lithium plating amount, and increase in side reaction resistance for the target battery.

7. The battery degradation mode prediction system according to claim 1, wherein, The lifetime data includes the degradation rate of the target battery for each degradation mode under specific cycles.

8. The battery degradation mode prediction system according to claim 7, wherein, The degradation rate is expressed as a percentage of the degradation assessment value of the target battery for each degradation mode relative to a reference value.

9. The battery degradation mode prediction system according to claim 1, wherein, The degradation mode analysis unit is configured to compare the degradation degree assessment value of the target battery for each degradation mode with a threshold based on the distribution data of the target battery for each degradation mode, and to visually output the degradation degree assessment value according to the comparison result between the degradation degree assessment value and the threshold.

10. The battery degradation mode prediction system according to claim 1, wherein, The degradation mode analysis unit is configured to identify degradation modes in which the degradation degree assessment value exceeds a threshold among the multiple degradation modes of the target battery based on the distribution data for each degradation mode of the target battery, and output the main degradation mode.

11. The battery degradation mode prediction system according to claim 1, wherein, The prediction model includes: The input layer is configured to receive the voltage distribution data for the target battery as input data; The intermediate layer is configured to generate, based on the input data, the distribution data of the target battery for each degradation mode as prediction data; and The output layer is configured to output the predicted data.

12. The battery degradation mode prediction system according to claim 11, wherein, The intermediate layer includes at least one of the following: The first intermediate layer is configured to extract features related to the voltage distribution data; The second intermediate layer is configured to process the continuous time series characteristics of the voltage distribution data; The third intermediate layer is configured to select weight information based on the correlation between the voltage distribution data and the distribution data for each degradation mode; and The fourth intermediate layer is configured to transform data dimensions based on information about the number of degradation patterns.

13. The battery degradation mode prediction system according to claim 12, wherein, The second intermediate layer is configured to receive the output of the first intermediate layer as input to the second intermediate layer.

14. A method for predicting battery degradation modes, comprising: Generate learning data, which includes a voltage distribution dataset for multiple reference cells and a distribution dataset for each degradation mode of the multiple reference cells; A prediction model is generated based on the learning data. The prediction model is trained to receive the voltage distribution dataset as input and output the distribution dataset corresponding to the voltage distribution dataset for each degradation mode. Obtain voltage distribution data for the target battery; The distribution data for each degradation mode of the target battery is obtained from the voltage distribution data by using the prediction model; as well as Based on the distribution data for each degradation mode, output lifetime data related to the degree of degradation of the target battery for each degradation mode. The distribution data for each degradation mode includes cycle-by-cycle degradation trend information for at least one of degradation parameters associated with the electrode plates of the target battery and degradation parameters associated with the resistance of the target battery.

15. The battery degradation mode prediction method according to claim 14, wherein, The voltage distribution dataset includes full-cell voltage distribution data for the plurality of reference cells, which is generated by an electrochemical model based on early lifetime data and half-cell data associated with the plurality of reference cells.

16. The battery degradation mode prediction method according to claim 15, wherein, The half-cell data includes positive electrode half-cell data and negative electrode half-cell data, and The full-cell voltage distribution data is generated based on the resistance characteristics extracted from each of the positive half-cell data and the negative half-cell data.

17. The battery degradation mode prediction method according to claim 14, wherein, The acquisition of the voltage distribution data for the target battery includes: The voltage value and charge / discharge capacity of the target battery are measured at predetermined sampling intervals; and Obtain the voltage distribution data, which includes the correlation between the voltage value and the charge / discharge capacity.

18. The battery degradation mode prediction method according to claim 14, wherein, The distribution data for each degradation mode includes information evaluated by the prediction model regarding at least one of cycle-by-cycle degradation of the positive electrode active material, degradation of the negative electrode active material, lithium plating amount, and increase in side reaction resistance for the target battery.

19. The battery degradation mode prediction method according to claim 14, wherein, The lifetime data includes the degradation rate of the target battery for each degradation mode under specific cycles.

20. A non-transitory computer-readable recording medium storing a program, which, when executed by a processor, causes the processor to perform the method according to any one of claims 14 to 19.

Images