System and method for predicting life of battery, and manufacturing process system of battery

Abstract

The invention relates to a system and method for predicting the life of a battery and a manufacturing process system of the battery. The system for predicting the life of a battery includes: a training data generation device configured to generate first data including life data of a reference battery and curve data of each battery degradation pattern; a prediction model generation means configured to generate, based on the first data, one or more life prediction models for predicting curve data for each battery degradation pattern using initial life data of the target battery as an input; and a life prediction device configured to predict a life of the target battery based on second data including curve data of each battery degradation pattern predicted by the one or more life prediction models based on initial life data of the target battery.

Description

Technical Field

[0001] The embodiments of this disclosure relate to systems and methods for predicting battery life and battery manufacturing process systems. Background Technology

[0002] When developing batteries such as rechargeable batteries, charging and discharging can be repeatedly performed under conditions and for periods similar to the actual usage environment of the battery, thus ensuring battery life. In this way, the long-term lifespan of the battery can be assessed by measuring its remaining capacity. Recently, with the increasing frequency of batteries being exposed to operating environments that accelerate degradation (e.g., fast charging, fast discharging, low or high temperature environments), more accurate prediction of degradation status is desired for battery safety and increased lifespan.

[0003] The information disclosed in this Background section is intended to enhance the understanding of the background of this disclosure, and therefore may contain information that does not constitute related (prior) art. Summary of the Invention

[0004] The problem this disclosure aims to solve is to provide a system and method for predicting battery life, as well as a battery manufacturing process system, for solving the aforementioned problems.

[0005] These and other aspects and features of this disclosure will be described in the following description of embodiments of this disclosure, or will become apparent from the following description of embodiments of this disclosure.

[0006] To address the aforementioned technical problems, a system for predicting battery life according to some embodiments of this disclosure may include a training data generation device configured to generate first data including lifespan data of a reference battery and curve data of each battery degradation mode; a prediction model generation device configured to generate one or more lifespan prediction models based on the first data for predicting curve data of each battery degradation mode using initial lifespan data of a target battery as input; and a lifespan prediction device configured to predict the lifespan of a target battery based on second data including curve data of each battery degradation mode predicted by the one or more lifespan prediction models based on the initial lifespan data of the target battery.

[0007] According to some embodiments, the prediction model generation apparatus may include a model training module configured to train one or more lifetime prediction models based on first data for predicting curve data of each battery degradation mode with reference battery lifetime data as input; and a degradation mode analysis module configured to determine a lifetime prediction model for verification corresponding to major degradation factors based on second data including curve data of each battery degradation mode output by one or more lifetime prediction models with initial lifetime data of the target battery as input.

[0008] According to some embodiments, the degradation mode analysis module can be configured to: select one or more models corresponding to the major degradation factors from one or more lifetime prediction models; and determine a lifetime prediction model for validation by combining the selected one or more models based on the weights of the major degradation factors.

[0009] According to some embodiments, the weights can be calculated based on the correlation between the main degradation factors and the lifespan of the target battery.

[0010] According to some embodiments, the lifetime prediction device can be configured to predict the state of health (SOH) of a target battery based on initial lifetime data of the target battery using a lifetime prediction model for validation.

[0011] According to some embodiments, the life prediction device can be configured to predict the health status based on the remaining life of the target battery, according to the life cycle of the target battery.

[0012] According to some embodiments, the life prediction device can be configured to predict the health status based on at least one of the target battery’s coulombic efficiency, charging time, voltage change, and charge slip / discharge slip.

[0013] According to some embodiments, one or more lifetime prediction models may include machine learning models based on convolutional neural networks (CNNs).

[0014] According to some embodiments, the first data may include curve data of each battery degradation mode generated by an electrochemical model based on lifetime start (BOL) data and semi-cell data for one or more batteries.

[0015] According to some embodiments, the second data may include at least one of the following: anode degradation, cathode degradation, active lithium consumption, and resistance increase of the target battery.

[0016] To address the aforementioned technical problems, a battery manufacturing process system according to some embodiments of this disclosure may include: a manufacturing apparatus configured to manufacture battery cells and inspect the quality of the manufactured battery cells based on the design specifications of the battery cells; and a life prediction system configured to predict the life of the battery cells by using one or more prediction models generated based on the design specifications of the battery cells, wherein the life prediction system is configured to predict the life of the battery cells based on curve data of each battery degradation mode predicted by one or more life prediction models based on the initial life data of the battery cells.

[0017] To address the aforementioned technical problems, a method for predicting battery life according to some embodiments of this disclosure may include generating first data including lifespan data of a reference battery and curve data of each battery degradation mode; generating one or more lifespan prediction models based on the first data for predicting curve data of each battery degradation mode using initial lifespan data of a target battery as input; generating second data including curve data of each battery degradation mode predicted by the one or more lifespan prediction models based on the initial lifespan data of the target battery; and predicting the lifespan of the target battery based on the generated second data.

[0018] According to some embodiments, the generation of the first data includes generating curve data for each battery degradation mode, generated by an electrochemical model based on BOL data and semi-cell data for one or more batteries.

[0019] According to some embodiments, one or more lifespan prediction models may include CNN-based machine learning models.

[0020] According to some embodiments, the second data may include at least one of the following: anode degradation, cathode degradation, active lithium consumption, and resistance increase of the target battery.

[0021] According to some embodiments, one or more lifetime prediction models may include one or more models among one or more lifetime prediction models that correspond to major degradation factors.

[0022] According to some embodiments, one or more models can be combined based on the weights of the main degradation factors.

[0023] According to some embodiments, the weights can be calculated based on the correlation between the main degradation factors and the lifespan of the target battery.

[0024] According to some embodiments, the prediction of the lifespan of a target battery may include: predicting the remaining lifespan of the target battery by one or more lifespan prediction models based on the lifespan cycle of the target battery.

[0025] According to some embodiments, the prediction of the lifetime of a target battery may include measuring at least one of the following: coulombic efficiency, charging time, voltage change, and charge slip / discharge slip of the target battery by one or more lifetime prediction models.

[0026] According to some embodiments of this disclosure, the lifespan of a battery can be predicted by analyzing the battery's degradation state in the early stages using initial lifespan data.

[0027] According to some embodiments of this disclosure, the resulting data on battery life prediction can be used in the development of battery cells.

[0028] However, the aspects and features of this disclosure are not limited to those described above, and those skilled in the art will clearly understand from the detailed description below that other aspects and features not mentioned will be apparent. Attached Figure Description

[0029] The accompanying drawings illustrate embodiments of the present disclosure, and together with the detailed description thereof, further describe aspects and features of the disclosure. Therefore, this disclosure should not be construed as limited to the drawings:

[0030] Figures 1 to 2 This is a block diagram describing a battery life prediction system according to some embodiments of the present disclosure.

[0031] Figure 3 This is a block diagram describing a predictive model generation apparatus according to some embodiments of the present disclosure.

[0032] Figure 4 This is a block diagram illustrating an information processing system for predicting battery life according to some embodiments of the present disclosure.

[0033] Figures 5A to 5C This is an example describing the input / output data of a lifetime prediction system according to a first embodiment of the present disclosure.

[0034] Figure 6 This is a flowchart describing a battery life prediction method according to some embodiments of the present disclosure.

[0035] Figure 7 and Figure 8 This is a block diagram describing a battery manufacturing process system according to some embodiments of the present disclosure.

[0036] Figure 9 It is a graph describing the procedures for predicting the lifespan and performing quality checks on a target battery performed by a battery manufacturing process system according to some other embodiments of this disclosure.

[0037] Explanation of some figure labels

[0038] 1: Manufacturing process system

[0039] 10: Life Prediction System

[0040] 100: Training data generation device

[0041] 200: Predictive model generation device

[0042] 210: Model Training Module

[0043] 220: Degradation Mode Analysis Module

[0044] 300: Life Prediction Device

[0045] 400: Information Processing System

[0046] 410: Memory

[0047] 420: Processor

[0048] 430: Communication Module

[0049] 440: Input / Output Interface

[0050] 1000: Manufacturing process equipment Detailed Implementation

[0051] In the following description, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Based on the principle that the inventor can be his / her own lexicographer to appropriately define the concepts of terms in order to best interpret his / her invention, the terms or words used in this specification and claims should not be construed as limited to their ordinary or dictionary meanings, but should be interpreted as having meanings and concepts consistent with the technical spirit of the present disclosure.

[0052] The embodiments described in this specification and the constructions shown in the accompanying drawings are merely some of the embodiments of this disclosure and do not represent all the technical ideas, aspects, and features of this disclosure. Accordingly, it should be understood that various equivalents and modifications that can replace or modify the embodiments described herein may exist at the time of filing this application.

[0053] It will be understood that when an element or layer is referred to as being "on," "connected to," or "coupled to" another element or layer, the element or layer may be directly on, directly connected to, or directly coupled to the other element or layer, or there may be one or more intermediary elements or layers. When an element or layer is referred to as being "directly on," "directly connected to," or "directly coupled to" another element or layer, there are no intermediary elements or layers. For example, when a first element is described as being "coupled to" or "connected to" a second element, the first element may be directly coupled to or directly connected to the second element, or the first element may be indirectly coupled to or indirectly connected to the second element via one or more intermediary elements.

[0054] In the figures, the dimensions of various elements, layers, etc., may be exaggerated for clarity. The same reference numerals designate the same elements. As used herein, the term “and / or” includes any and all combinations of one or more of the items listed. Furthermore, when describing embodiments of this disclosure, the use of “may” relates to “one or more embodiments of this disclosure.” Expressions such as “at least one of” and “any one of”, when placed after a list of elements, modify the entire list of elements without modifying individual elements in that 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 specify 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 term “use” and its variations may be considered synonymous with the term “utilize” and its variations, respectively. As used herein, the terms “substantially,” “approximately,” and similar terms are used as approximate terms rather than terms of degree and are intended to describe the inherent biases of measurements or calculations that would be recognized by one of ordinary skill in the art.

[0055] It will be understood that although the terms first, second, third, etc., are used herein to describe various elements, components, regions, layers, and / or portions, these elements, components, regions, layers, and / or portions should not be limited by these terms. These terms are used to distinguish one element, component, region, layer, or portion from another element, component, region, layer, or portion. Therefore, the first element, component, region, layer, or portion discussed below may be referred to as the second element, component, region, layer, or portion without departing from the teachings of the exemplary embodiments.

[0056] In this document, for ease of description, spatial relative terms such as “below,” “under,” “below,” “above,” and “above” are used to describe the relationship between one element or feature illustrated in the figures and another element(s). It will be understood that, in addition to the orientation 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, the element described as “below” or “below” other elements or features will then be oriented “above” or “above” that other element or feature. Thus, the term “below” can encompass both the orientations of above and below. The device may be oriented in other ways (rotated 90 degrees or otherwise), and the spatial relative descriptive terms used herein should be interpreted accordingly.

[0057] 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," "including," and / or variations thereof specify the presence of 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.

[0058] Furthermore, any numerical range disclosed and / or described herein is intended to include all subranges with the same numerical precision that fall 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), i.e., subranges such as 2.4 to 7.6 that have a minimum value equal to or greater than 1.0 and a maximum value equal to or less than 10.0. Any maximum numerical limit described herein is intended to include all smaller numerical limits that fall within it, and any minimum numerical limit described herein is intended to include all larger numerical limits that fall within it. Accordingly, the applicant reserves the right to amend this specification (including the claims) to explicitly describe any subranges that fall within the range explicitly described herein.

[0059] 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 a deviation considered low in the art (e.g., 5% or less). Furthermore, when a parameter is said to be consistent in a given region, this can mean that it is consistent in terms of average value.

[0060] Throughout this specification, unless otherwise stated, each element may be singular or plural.

[0061] Placing any element "above (or below)" or "above (or below)" another element can mean that the arbitrary element can be positioned in contact with the upper (or lower) surface of the other element, and that another element can also be positioned between the other element and the arbitrary element positioned above (or below) the other element.

[0062] Furthermore, it will be understood that when a component is referred to as “linked,” “coupled,” or “connected” to another component, these components can be directly “coupled,” “linked,” or “connected” to each other, or another component can be “between” these components.

[0063] Throughout this specification, unless otherwise stated, when “A and / or B” is mentioned, it means A, B, or A and B. That is, “and / or” includes any or all combinations of the enumerated items. Unless otherwise stated, when “C to D” is mentioned, it means C or more and D or fewer.

[0064] Figure 1 and Figure 2 This is a block diagram illustrating a battery life prediction system 10 according to some embodiments of the present disclosure. Figure 3 This is a block diagram illustrating a predictive model generation apparatus 200 according to some embodiments of the present disclosure.

[0065] refer to Figure 1 and Figure 2 The lifespan prediction system 10 may include a training data generation device 100, a prediction model generation device 200, and a lifespan prediction device 300. Furthermore, the lifespan prediction system 10 may further include an input device for receiving various data required to generate training data, an output device for outputting lifespan prediction results (e.g., the health status of a target battery), a storage device for storing training data, lifespan prediction models, etc., and a communication device for communicating with external devices.

[0066] According to existing technology, in order to analyze the degradation factors of a battery cell, the ratio of voltage change to charging capacity change (dV / dQ) is calculated based on the voltage curve, or the battery cell is disassembled. However, this analysis method is not only complex but also time-consuming. A life prediction system 10 according to some embodiments of this disclosure can estimate the degradation factors of a battery cell and predict its lifespan by inputting the initial lifespan data of the battery cell into the life prediction system 10.

[0067] In the following description, "reference battery" can refer to a battery capable of generating data for training a lifespan prediction model. Additionally, "target battery" can refer to a battery intended for predicting health status, etc., using a lifespan prediction model. 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.

[0068] In some embodiments, the training data generation apparatus 100 may generate first data including lifetime data of a reference battery and curve data for each battery degradation mode. For example, the lifetime data of the reference battery may include data such as charging capacity and voltage measured during a predetermined lifetime cycle of the reference battery. Here, the full lifetime cycle of the reference battery may refer to the lifetime cycle from the BOL (Start of Life) to the EOL (End of Life) of the reference battery. For example, the lifetime data may be lifetime data measured over 1000 lifetime cycles. However, this disclosure is merely an example and is not limited thereto, and the lifetime data may vary in the range of the predetermined lifetime cycle depending on the type of reference battery.

[0069] Furthermore, the first data may include curve data for each battery degradation mode. For example, curve data for each battery degradation mode may include, but is not limited to, anodic degradation curve data, cathode degradation curve data, active lithium consumption curve data, resistance increase curve, etc. Curve data for each battery degradation mode can be generated from the battery's electrochemical model. For example, curve data for each battery degradation mode can be generated by inputting BOL (Balance of Life) data and semi-cell data for one or more batteries into the battery's electrochemical model. Semi-cell data is data measured from a semi-cell configured for use when testing battery cells, and may include anodic and cathode cell data. The training data generation device 100 can input BOL data and semi-cell data into the electrochemical model, and curve data for each battery degradation mode can be generated by the electrochemical model. Here, curve data for each battery degradation mode may include information about at least one internal structure of the battery or factors affecting battery degradation, and can be defined according to the battery's electrochemical model.

[0070] In some embodiments, the prediction model generation apparatus 200 can generate one or more lifetime prediction models trained to predict the curve data of each battery degradation mode of the target battery, based on first data including lifetime data of a reference battery and curve data of each battery degradation mode. For example, the prediction model generation apparatus 200 can generate a lifetime prediction model by training it with the correlation between lifetime data of a predetermined lifetime cycle of the reference battery and the curve data of each battery degradation mode corresponding thereto. Additionally, the prediction model generation apparatus 200 can generate a verifiable lifetime prediction model corresponding to major degradation factors based on second data including curve data of each battery degradation mode output by one or more lifetime prediction models using initial lifetime data of the target battery as input. For example, the prediction model generation apparatus 200 can determine the major degradation factors of the battery based on the second data using a degradation factor analysis algorithm. In this way, the verifiable lifetime prediction model can be generated based on the determined major degradation factors.

[0071] refer to Figure 1 and Figure 3 The prediction model generation device 200 may include a model training module 210 and a degradation mode analysis module 220.

[0072] In some embodiments, the model training module 210 can train one or more lifetime prediction models to predict curve data for each battery degradation mode using initial lifetime data of a specific type of battery as input. For example, the model training module 210 can train one or more lifetime prediction models using first data. For example, the model training module 210 can train a lifetime prediction model to predict anode degradation curve data using initial lifetime data of a target battery as input, based on training data including lifetime data of a reference battery and its corresponding anode degradation curve data. Similarly, the model training module 210 can train a lifetime prediction model to predict cathode degradation curves using initial lifetime data of a target battery as input, based on training data including lifetime data of a reference battery and its corresponding cathode degradation curve data. In this way, the model training module 210 can generate one or more lifetime prediction models trained for each battery degradation mode.

[0073] Each of one or more lifespan prediction models may include a machine learning model based on a convolutional neural network (CNN). However, this disclosure is not limited thereto.

[0074] In some embodiments, the degradation mode analysis module 220 may determine, based on second data, a verifiable lifetime prediction model corresponding to the main degradation factors. Here, the second data may include curve data of each battery degradation mode output by one or more lifetime prediction models with initial lifetime data of the target battery as input. For a specific example, the second data may include at least one or more of anode degradation, cathode degradation, active lithium consumption, and resistance increase based on the target battery's lifetime cycles (e.g., 100 initial lifetime cycles), but this disclosure is not limited thereto. That is, the second data may include basic data that enables the identification of battery degradation factors based on the target battery's lifetime cycles. Here, battery degradation factors may be data related to at least one internal state of the battery that changes during the initial lifetime cycles.

[0075] The degradation mode analysis module 220 can select one or more models corresponding to the major degradation factors from one or more lifetime prediction models based on second data. Here, the major degradation factor can refer to the degradation factor that has the greatest impact on the health status of the target battery among multiple degradation factors. Specifically, the degradation mode analysis module 220 can determine one or more of the following as major degradation factors based on the amount of anode degradation, cathode degradation, active lithium consumption, and resistance increase (side reaction) over the lifetime of the target battery: anode degradation, cathode degradation, active lithium consumption, and resistance increase (side reaction). Furthermore, the degradation mode analysis module 220 can assign weights to each of the one or more major degradation factors proportionally to their impact on the lifetime of the target battery. The degradation mode analysis module 220 can combine one or more lifetime prediction models corresponding to the one or more major degradation factors based on the weights assigned to them. Here, the weights can be calculated based on the correlation between the major degradation factors and the lifetime of the target battery. The weights can be determined based on the correlation between the lifetime of the target battery and the degradation factors, which is calculated using an appropriate degradation factor analysis algorithm such as correlation analysis.

[0076] For example, the degradation mode analysis module 220 can identify the amount of anode degradation as the primary degradation factor with the greatest impact on the lifespan of the target battery. In this case, the degradation mode analysis module 220 can determine a lifespan prediction model for verification based on curve data that calculates the amount of anode degradation using the initial lifespan data of the target battery as input. Alternatively, the degradation mode analysis module 220 can assign the maximum weight to a lifespan prediction model that calculates the amount of anode degradation using curve data that calculates the amount of anode degradation using the initial lifespan data of the target battery as input. Furthermore, the degradation mode analysis module 220 can assign lower weights to degradation factors other than the amount of anode degradation. By combining one or more corresponding lifespan prediction models using one or more weights thus determined, a lifespan prediction model for verification can be generated.

[0077] For example, the degradation mode analysis module 220 can identify cathode degradation as the primary degradation factor with the greatest impact on the lifespan of the target battery, and resistance increase as the secondary degradation factor with the second greatest impact. In this case, the degradation mode analysis module 220 can assign the maximum weight to cathode degradation and assign weights exceeding a threshold to resistance increase that are smaller than those assigned to cathode degradation. Accordingly, the degradation mode analysis module 220 can select from one or more lifespan prediction models the lifespan prediction model associated with cathode degradation and the lifespan prediction model associated with resistance increase.

[0078] The degradation mode analysis module 220 can determine a lifetime prediction model for validation based on weights. For example, if the degradation mode analysis module 220 identifies anode degradation as the primary degradation factor, it can identify a lifetime prediction model associated with anode degradation as the lifetime prediction model for validation. In other examples, if the degradation mode analysis module 220 identifies cathode degradation and resistance increase as the primary degradation factors, it can identify a lifetime prediction model associated with cathode degradation and resistance increase, or a combination of these models, as the lifetime prediction model for validation.

[0079] Then, the degradation mode analysis module 220 can transmit the determined lifetime prediction model for verification to the lifetime prediction device 300.

[0080] Refer again Figure 1 and Figure 2 The lifespan prediction device 300 can predict the lifespan of a target battery based on its initial lifespan data. For example, the lifespan prediction device 300 can predict the state of health (SOH) of the target battery by inputting the initial lifespan data into a lifespan prediction model for verification. Specifically, the lifespan prediction device 300 can calculate curve data for each degradation mode by inputting the initial lifespan data into one or more lifespan prediction models included in the lifespan prediction models for verification. Furthermore, the lifespan prediction device 300 can calculate the remaining lifespan by inputting the calculated curve data for each degradation mode into an appropriate function or algorithm for calculating the battery's aging rate. The lifespan prediction device 300 can predict the state of health based on the remaining lifespan of the target battery according to the calculated lifespan cycle of the target battery.

[0081] Furthermore, the lifetime prediction device 300 can calculate at least one of the following for the target battery: coulombic efficiency, charging time, voltage change, and charge / discharge slip, based on curve data for each degradation mode calculated by a lifetime prediction model for validation. Additionally, the lifetime prediction device 300 can predict the health status of the target battery based on at least one or more of the following: coulombic efficiency, charging time, voltage change, and charge / discharge slip. Here, charge / discharge slip can refer to the change in capacity in the accumulated capacity-voltage curve data. For example, charge / discharge slip can include both charge slip and discharge slip. Charge slip can refer to the change in charged capacity in the accumulated capacity-voltage curve data, and discharge slip can refer to the change in discharged capacity in the accumulated capacity-voltage curve data.

[0082] As described above, the battery life prediction system 10 according to some embodiments of this disclosure can predict the battery life by analyzing the battery's degradation state in the early stages using initial battery life data. Furthermore, the life prediction system 10 can use the predicted battery life data for the development of individual battery cells.

[0083] Figure 4 This is a block diagram illustrating an information processing system 400 for predicting battery life according to some embodiments of the present disclosure.

[0084] For example, information processing system 400 can be with Figure 1 and Figure 2 This corresponds to the battery life prediction system 10 shown. The information processing system 400 may include a memory 410, a processor 420, a communication module 430, and an input / output interface 440. (Reference) Figure 4 The information processing system 400 can be configured to transmit information and / or data over a network using the communication module 430. According to some embodiments, the information processing system 400 may be formed of at least one device including a memory 410, a processor 420, a communication module 430, and an input / output interface 440.

[0085] Memory 410 may include any non-transitory computer-readable recording medium. According to some embodiments, memory 410 may include a permanent mass storage device such as read-only memory (ROM), a disk drive, a solid-state drive (SSD), flash memory, etc. As another example, a permanent mass storage device such as ROM, SSD, flash memory, disk drive, etc., may be included in the information processing system 400 as a separate permanent storage device, distinct from memory 410. Additionally, memory 410 may store software components including an operating system and at least one program code (e.g., code for generating and training a lifespan prediction model installed in and running in the information processing system 400, and the lifespan prediction model predicting the health status of a target battery).

[0086] Software components can be loaded from a computer-readable recording medium separate from memory 410. This separate computer-readable recording medium may include a recording medium that can be directly connected to information processing system 400, and may include, for example, computer-readable recording media such as floppy disk drives, magnetic disks, magnetic tapes, DVD / CD-ROM drives, memory cards, etc. As another example, software components may be loaded into memory 410 via communication module 430 instead of a computer-readable recording medium. For example, at least one program may be loaded onto memory 410 based on a computer program (e.g., a program for generating and training a lifespan prediction model, a lifespan prediction model predicting the health state of a target battery, etc.) provided via communication module 430 by a file distribution system that distributes installation files for applications.

[0087] Processor 420 can be configured to process commands of a computer program by performing basic arithmetic, logic, and input / output operations. Commands can be provided to a user terminal (not shown) or another external system by memory 410 or communication module 430. For example, processor 420 can receive lifetime data of a reference battery and / or initial lifetime data of a target battery, lifetime data of various types of batteries, etc., from one or more external devices or battery manufacturing process apparatus. It can generate and train a lifetime prediction model based on the lifetime data of the reference battery and curve data for each degradation mode, or it can calculate the health status of the target battery based on the initial lifetime data of the target battery and the lifetime prediction model.

[0088] The communication module 430 can provide the user terminal (not shown) and the information processing system 400 with the configuration or function to communicate with each other via a network, and can also provide the information processing system 400 with the configuration or function to communicate with external systems (for example, a separate battery manufacturing process device, a cloud system, etc.). For example, control signals, commands, data, etc., provided under the control of the processor 420 of the information processing system 400 can be transmitted to the user terminal and / or the external system via the communication module 430 and the network through the communication module 430. For instance, the health status of the target battery generated by the information processing system 400 can be transmitted to the user terminal and / or the external system via the communication module 430 and the network through the communication module 430. Furthermore, the user terminal and / or the external system that has received the health status of the target battery, etc., can output the received information via a device with display output capability.

[0089] Furthermore, the input / output interface 440 of the information processing system 400 can be a device for engaging with an input or output device (not shown) that can be connected to the information processing system 400 or that may be included in the information processing system 400. Figure 4 In this diagram, the input / output interface 440 is shown as an element configured separately from the processor 420, but is not limited thereto, and the input / output interface 440 may be configured to be included within the processor 420. The information processing system 400 may include more than Figure 4 The components in the text are more numerous. However, it is not necessary to explicitly show most of the components in the existing technology.

[0090] The processor 420 of the information processing system 400 can be configured to manage, process, and / or store information and / or data received from multiple user terminals and / or multiple external systems. According to some embodiments, the processor 420 can receive lifespan data of a reference battery and / or initial lifespan data of a target battery, lifespan data of various types of batteries, etc., from user terminals and / or external systems. The processor 420 can calculate the health status of the target battery based on the initial lifespan data of the target battery using a lifespan prediction model, and output the calculated health status, etc., via a device connected to the information processing system 400 with display output capability.

[0091] Figures 5A to 5C This is an example describing the input / output data of the lifetime prediction system 10 according to a first embodiment of the present disclosure.

[0092] Figure 5A This is a graph showing the charging capacity based on the lifespan of the target battery cell. Specifically, Figure 5AThis is a graph showing the change in charge capacity measured during the initial life cycle for each of the four Designs of Experiments (DOEs) for a target battery. Figure 5A The initial lifetime data of the target battery cell shown can be input into the lifetime prediction system 10. For example, the charging capacity data corresponding to the initial lifetime cycle of the target battery cell can be input into the lifetime prediction model of the lifetime prediction system 10 for verification.

[0093] Figure 5B and Figure 5C show when Figure 5A The curves showing the amount of cathode degradation and resistance increase (side reaction) are calculated when the initial lifetime data of the target battery is input into the lifetime prediction model for verification. For example, the degradation mode analysis module 220 of the lifetime prediction system 10 can identify the amount of cathode degradation and resistance increase as the main degradation factors of the target battery, and use two corresponding lifetime prediction models to generate a lifetime prediction model for verification. Accordingly, the lifetime prediction model for verification can use the initial lifetime data of the target battery as input to calculate the curves of cathode degradation and resistance increase.

[0094] Figure 6 This is a flowchart describing a battery life prediction method 600 according to some embodiments of the present disclosure. Figure 6 The lifespan prediction method 600 can be used Figure 1 and Figure 2 The life prediction system 10 is used to perform this.

[0095] refer to Figure 6 The life prediction method 600 can begin by generating first data (step 610) that includes life data of a reference battery and curve data of each battery degradation mode.

[0096] According to some embodiments, generating the first data (step 610) may include generating curve data for each battery degradation mode generated by an electrochemical model based on BOL data and semi-cell data for one or more batteries.

[0097] Additionally, based on the first data, one or more lifetime prediction models can be generated to predict the degradation mode of each battery using the initial lifetime data of the target battery as input (step 620).

[0098] According to some embodiments, one or more lifetime prediction models may include machine learning models based on convolutional neural networks (CNNs).

[0099] Then, a second set of data can be generated, which includes curve data of each battery degradation mode predicted by one or more lifetime prediction models based on the initial lifetime data of the target battery (step 630).

[0100] According to some embodiments, the second data may include at least one or more of the following: anode degradation, cathode degradation, active lithium consumption, and resistance increase of the target battery. Additionally, one or more lifetime prediction models may include one or more models corresponding to the major degradation factors. The one or more models may be combined based on the weights of the major degradation factors. Here, the weights may be calculated based on the correlation between the major degradation factors and the lifetime of the target battery.

[0101] The lifespan of the target battery can be predicted based on the second data (step 640).

[0102] According to some embodiments, predicting the lifespan of the target battery (step 640) may include predicting the remaining lifespan of the target battery by one or more lifespan prediction models based on the lifespan cycle of the target battery. In other embodiments, predicting the lifespan of the target battery (step 640) may include predicting at least one or more of the following by one or more lifespan prediction models: coulombic efficiency, charging time, voltage change, and charge slip / discharge slip.

[0103] Figure 7 and Figure 8 This is a block diagram describing a battery manufacturing process system 1 according to some embodiments of the present disclosure. Figure 7 and Figure 8 The life prediction system 10 can be combined with Figure 1 and Figure 2 This corresponds to the lifetime prediction system 10 described herein. Therefore, in the following description of the lifetime prediction system 10, repeated descriptions of its construction and function will be omitted.

[0104] refer to Figure 7 and Figure 8 The manufacturing process system 1 may include a life prediction system 10 and a manufacturing process apparatus 1000.

[0105] refer to Figure 7The lifespan prediction system 10 can predict the lifespan of a battery cell using one or more prediction models generated based on the design specifications of the battery cell. Specifically, the lifespan prediction system 10 can train and generate one or more lifespan prediction models using first data, which includes curve data of each battery degradation mode generated using an electrochemical model of the battery based on the design specifications of the battery cell and lifespan data of a reference battery. Furthermore, the lifespan prediction system 10 can calculate the health state of a target battery by inputting the initial lifespan data of the target battery cell into one or more lifespan prediction models.

[0106] In some embodiments, the manufacturing apparatus 1000 can manufacture battery cells based on the design specifications of the battery cells. Additionally, the manufacturing apparatus 1000 can inspect the quality of the manufactured battery cells based on lifespan prediction results, including health status, calculated by the lifespan prediction system 10.

[0107] refer to Figure 8 The manufacturing process apparatus 1000 can perform battery cell manufacturing processes and quality inspection processes. For example, the battery cell manufacturing process may include a coating process 1010, a rolling process 1020, a cutting and grooving process 1030, a battery cell assembly and injection process 1040, an activation and aging process 1050, and a degassing process 1060. Additionally, the battery cell quality inspection process may include, for example, a quality inspection 1070 and a shipment 1080.

[0108] Specifically, coating process 1010 can be a process of coating a slurry containing active materials such as cathodes or anodes onto the current collector of a battery cell. Rolling process 1020 can be a process of flatly rolling the electrodes of a battery cell coated with active materials. Cutting and grooving process 1030 can be a process of cutting electrodes to fit the size of the battery cell and forming and processing electrode terminals, etc. Battery cell assembly and injection process 1040 can include a process of assembling a battery cell by inserting an electrode assembly including a positive electrode plate and a negative electrode plate into a housing, sealing it with a cover assembly, etc., and then injecting electrolyte into the battery cell. Activation and aging process 1050 can be a process of stabilizing a battery cell by charging and discharging the battery cell. Degassing process 1060 can be a process of removing gases generated in activation and aging process 1050 from inside the battery cell.

[0109] In some embodiments, if a new target battery cell is developed at the request of a customer, the life prediction system 10 can receive data on the specifications of the target battery cell. Additionally, the target battery cell can be manufactured by the manufacturing process apparatus 1000 according to specifications. When performing a quality check 1070 on the target battery cell, the manufacturing process apparatus 1000 can transmit the initial life data of the target battery cell to the life prediction system 10. The life prediction system 10 can generate a life prediction model based on the curve data of each battery degradation mode according to the specifications of the target battery cell, and predict the life of the target battery cell based on the initial life data using the life prediction model. The life prediction system 10 can then transmit the predicted life and major degradation factors of the target battery cell to the manufacturing process apparatus 1000. Accordingly, the manufacturing process apparatus 1000 can determine whether the predicted life of the target battery cell meets the customer's requirements during the quality check 1070. If the target battery cell does not meet the customer's requirements, the manufacturing process apparatus 1000 can remanufacture the target battery cell by changing or adjusting various manufacturing process parameters used in the manufacturing process of the target battery cell based on the received major degradation factors. Conversely, if the target battery cell meets the customer's requirements, the manufacturing process unit 1000 can ship (1080) the target battery cell. (The sentence is incomplete and requires further context.) Figure 9 The document provides a detailed description of a specific example of manufacturing process system 1.

[0110] Figure 9 It is a graph describing the procedures for predicting the lifespan and performing quality checks on a target battery performed by a battery manufacturing process system 1 according to some other embodiments of the present disclosure.

[0111] refer to Figure 9 Curve X in the graph can be the initial lifetime data of the target battery cell. Here, the initial lifetime data of the target battery cell can be the state of health (SOH) in the range of 1 lifetime cycle to 100 lifetime cycles. However, this disclosure is not limited thereto.

[0112] Curve Y in the graph shown can be the lifetime of a target battery cell predicted by the lifetime prediction system 10 according to some embodiments of this disclosure. Curve E can be the lifetime of a target battery cell measured through actual experiments.

[0113] As a specific example, when the life cycle of the target battery cell is C_N1, the customer who requested the development of the target battery cell can require that the corresponding state of health (SOH) of the battery be 90%. Additionally, the customer can require that the life cycle in which the target battery cell exhibits a sudden drop in its state of health be C_N2.

[0114] During the quality inspection of the target battery, the manufacturing process system 1 can check the conditions requested by the customer by utilizing the life status of the target battery calculated by the life prediction system 10.

[0115] refer to Figure 9 Referring to curve Y, which represents the output lifespan state of the lifespan prediction system 10, it can be confirmed that when the state of health (SOH) of the target battery cell is 90% (N1), the lifespan of the target battery cell is C1. In comparison, referring to curve E, which represents the actual experimental curve, it can be confirmed that the lifespan of the target battery cell is C2. That is, it can be confirmed that the manufactured target battery cell exhibits a better lifespan than the customer's requirements. Furthermore, it can be confirmed that the result obtained by predicting the lifespan of the target battery cell using the lifespan prediction system 10 has a very small difference from the actual measured lifespan of the target battery. In other words, the accuracy of the lifespan prediction system 10 is confirmed to be 90% or higher by the shown graphs.

[0116] Furthermore, referring to curve Y, which is the output curve of the life prediction system 10 shown in the graph, it can be confirmed that the life cycle of the target battery cell is C3 when the state of health (SOH) of the target battery cell suddenly drops (N2). In comparison, referring to curve E, which is the actual experimental curve, it can be confirmed that the life cycle of the target battery cell is C4. That is to say, it can be confirmed that the manufactured target battery cell exhibits a better life cycle than the customer's requirements. In addition, the graph shown confirms that the result obtained by predicting the life cycle of the target battery cell by the life prediction system 10 does not differ significantly from the actual measured life cycle of the target battery.

[0117] As described above, the battery life prediction system 10 according to some embodiments of this disclosure can predict the battery life by analyzing the battery's degradation state in the early stages using initial battery life data. Furthermore, the life prediction system 10 can use the predicted battery life data for the development and manufacturing processes of individual battery cells.

[0118] Although this disclosure has been described with reference to the accompanying drawings illustrating aspects of the present disclosure, this disclosure is not limited thereto. Those skilled in the art to which this disclosure pertains will be able to make various modifications and variations within the spirit and scope of the present disclosure, the claims, and their equivalents.

Claims

1. A system for predicting battery life, comprising: The training data generation device is configured to generate first data including lifetime data of a reference battery and curve data of each battery degradation mode; A prediction model generation apparatus is configured to, based on the first data, generate one or more lifetime prediction models for predicting curve data of each battery degradation mode using initial lifetime data of the target battery as input; and The life prediction device is configured to predict the life of the target battery based on second data including curve data of each battery degradation mode predicted by the one or more life prediction models based on the initial life data of the target battery.

2. The system according to claim 1, wherein, The prediction model generation device includes: A model training module is configured to train, based on the first data, one or more lifetime prediction models for predicting curve data of each battery degradation mode using the lifetime data of the reference battery as input; and The degradation mode analysis module is configured to determine a verifiable lifetime prediction model corresponding to the major degradation factors based on the second data, which includes curve data of each battery degradation mode output by the one or more lifetime prediction models with the initial lifetime data of the target battery as input.

3. The system according to claim 2, wherein, The degradation mode analysis module is configured as follows: Select one or more models from the one or more lifetime prediction models that correspond to the major degradation factors; and The lifetime prediction model for validation is determined by combining one or more selected models based on the weights of the main degradation factors.

4. The system according to claim 3, wherein, The weights are calculated based on the correlation between the major degradation factors and the lifespan of the target battery.

5. The system according to claim 2, wherein, The lifetime prediction device is configured to predict the health status of the target battery based on the initial lifetime data of the target battery using the verifiable lifetime prediction model.

6. The system according to claim 5, wherein, The lifespan prediction device is configured to predict the health status based on the remaining lifespan of the target battery, according to the lifespan cycle of the target battery.

7. The system according to claim 6, wherein, The life prediction device is configured to predict the health status based on at least one of the target battery's coulombic efficiency, charging time, voltage change, and charge slip / discharge slip.

8. The system according to any one of claims 1 to 7, wherein, The one or more lifetime prediction models include machine learning models based on convolutional neural networks (CNNs).

9. The system according to any one of claims 1 to 7, wherein, The first data includes the curve data for each battery degradation mode generated by an electrochemical model based on lifetime start-of-life (BOL) data and semi-cell data for one or more cells.

10. The system according to any one of claims 1 to 7, wherein, The second data includes at least one of the following: anode degradation, cathode degradation, active lithium consumption, and resistance increase of the target battery.

11. A battery manufacturing process system, comprising: A manufacturing apparatus is configured to manufacture battery cells and inspect the quality of the manufactured battery cells based on the design specifications of the battery cells. as well as A lifespan prediction system is configured to predict the lifespan of the battery cell using one or more prediction models generated based on the design specifications of the battery cell. The life prediction system is configured to predict the life of the battery cell based on curve data of each battery degradation mode predicted by the one or more life prediction models based on the initial life data of the battery cell.

12. A method for predicting battery life, comprising: Generate first data including lifespan data of the reference battery and curve data of each battery degradation mode; Based on the first data, one or more lifetime prediction models are generated to predict the curve data of each battery degradation mode using the initial lifetime data of the target battery as input. Generate second data including curve data for each battery degradation mode predicted by the one or more lifetime prediction models based on the initial lifetime data of the target battery; and The lifespan of the target battery is predicted based on the generated second data.

13. The method according to claim 12, wherein, The generation of the first data includes: The curve data for each battery degradation mode is generated by an electrochemical model based on BOL data and semi-cell data for one or more batteries.

14. The method according to claim 12, wherein, The one or more lifespan prediction models include CNN-based machine learning models.

15. The method according to claim 12, wherein, The second data includes at least one of the following: anode degradation, cathode degradation, active lithium consumption, and resistance increase of the target battery.

16. The method according to claim 12, wherein, The one or more lifetime prediction models include one or more models among the one or more lifetime prediction models that correspond to the major degradation factors.

17. The method according to claim 16, wherein, The one or more models are combined based on the weights of the main degradation factors.

18. The method according to claim 17, wherein, The weights are calculated based on the correlation between the major degradation factors and the lifespan of the target battery.

19. The method of claim 16, wherein, The prediction of the lifespan of the target battery includes: The remaining lifespan of the target battery is predicted by the one or more lifespan prediction models based on the lifespan cycle of the target battery.

20. The method of claim 16, wherein, The prediction of the lifetime of the target battery includes: The target battery is predicted by one or more lifetime prediction models, including at least one of the following: coulombic efficiency, charging time, voltage change, and charge / discharge slip.

Images