Server and its battery analysis method

Abstract

The present invention provides a battery analysis method and a server for performing the same. The battery analysis method includes the steps of: performing electrochemical impedance spectroscopy (EIS) on a battery cell at a plurality of different states of charge to obtain EIS information for the plurality of states of charge; calculating a distribution of relaxation times (DRT) for the EIS information for the plurality of states of charge to generate DRT information for the plurality of states of charge; and generating a characteristic image corresponding to the battery cell, including information on the magnitude of impedance as a function of frequency calculated for the plurality of states of charge, based on the DRT information for the plurality of states of charge.

Description

[Technical Field]

[0001] This application claims the benefit of priority based on Korean Patent Application No. 2023-0174550 dated December 5, 2023 and Korean Patent Application No. 2023-0174544 dated December 5, 2023, and all contents disclosed in the documents of the relevant Korean patent applications are incorporated herein by reference.

[0002] The present disclosure relates to a method for analyzing a battery and a server for performing the method. [Background technology]

[0003] One of the conventional techniques used to analyze battery cells is electrochemical impedance spectroscopy (EIS). This is an analytical method in which an AC voltage or current at various frequencies is applied to a battery cell, and the impedance is calculated based on the resulting measured AC current or voltage, and this is then expressed as a Nyquist plot. This has been widely used because it allows for non-destructive analysis of battery cells.

[0004] One of the conventional techniques used to more easily analyze such EIS results is the Distribution of Relaxation Times (DRT), which is widely used because it does not require prior knowledge of the battery's impedance and can convert the impedance data calculated as a result of EIS into a distribution of relaxation times to identify the equivalent circuit of the battery.

[0005] However, the analysis results of the battery obtained as a result of EIS and DRT still only show somewhat limited characteristics of the battery, and therefore, an analytical method capable of analyzing more diverse characteristics of the battery is needed. Summary of the Invention [Problem to be solved by the invention]

[0006] The disclosed embodiments provide a battery analysis method and a server for performing the same. Specifically, an object of the present invention is to provide a method for analyzing various battery characteristics based on the results of EIS and DRT. Another object of the present invention is to provide a method for deriving various battery characteristics by analyzing characteristic images of battery cells obtained based on the results of EIS and DRT.

[0007] The technical objectives to be achieved by the present embodiment are not limited to those described above, and other technical objectives can be inferred from the following embodiments. [Means for solving the problem]

[0008] One aspect of the present disclosure may provide a battery analysis method performed on a server, the method including: performing electrochemical impedance spectroscopy (EIS) on a battery cell at a plurality of different states of charge to obtain EIS information for the plurality of states of charge; calculating a distribution of relaxation times (DRT) for the EIS information for the plurality of states of charge to generate DRT information for the plurality of states of charge; and generating a feature image corresponding to the battery cell, including information on impedance magnitude according to frequency calculated for the plurality of states of charge, based on the DRT information for the plurality of states of charge.

[0009] In one embodiment of the present disclosure, the battery analysis method may include generating the feature image in the form of a two-dimensional image determined based on the impedance magnitude information, where a first axis is the frequency, a second axis is the state of charge, and a shading level of each pixel is calculated for each of the frequencies and the plurality of states of charge, based on the DRT information for each of the plurality of states of charge.

[0010] In one embodiment of the present disclosure, the generating of the characteristic image may include the steps of: determining first magnitude information of impedance according to the frequency for each of the plurality of states of charge for a charging process of the battery cell based on first DRT information for each of the plurality of states of charge during the charging process of the battery cell, and generating a first characteristic image for the charging process, which is at least a part of the characteristic images, based on the first magnitude information; and determining second magnitude information of impedance according to the frequency for each of the plurality of states of charge for a discharging process of the battery cell based on second DRT information for each of the plurality of states of charge during the discharging process of the battery cell, and generating a second characteristic image for the discharging process, which is at least a part of the characteristic images, based on the second magnitude information.

[0011] Also, in one embodiment of the present disclosure, the battery analysis method may include the feature image including the first and second feature images symmetrically about a reference axis parallel to the first axis.

[0012] In addition, an embodiment of the present disclosure may include a battery analysis method further including storing a set of characteristic images for a plurality of reference battery cells, including the characteristic image, in a database.

[0013] In addition, in one embodiment of the present disclosure, a battery analysis method may be included in which the characteristic image set is stored in association with at least a portion of reference characteristic information including at least a portion of material information, deterioration level information, and usage environment information of each reference battery cell corresponding to each image included in the characteristic image set.

[0014] In addition, in one embodiment of the present disclosure, the battery analysis method may further include the steps of: performing the EIS on a target battery cell at a plurality of different charge states to obtain a plurality of target EIS information, calculating the DRT for the target EIS information to obtain target DRT information, and generating a target image for the target battery cell based on the target DRT information; and estimating feature information of the target battery cell based on the feature image set and the target image.

[0015] In addition, in one embodiment of the present disclosure, the step of estimating the feature information of the target battery cell may include a step of calculating a similarity between the target image and an image included in the feature image set, and then estimating reference feature information associated with the image having the highest similarity as the feature information of the target battery cell.

[0016] In addition, in one embodiment of the present disclosure, the step of estimating the feature information of the target battery cell may include a battery analysis method including inputting the target image to a neural network trained by inputting the feature image set and reference feature information corresponding to each image included therein as learning data and ground truth information therefor, and processing the target image through the neural network to estimate output feature information as the feature information of the target battery cell.

[0017] In addition, in one embodiment of the present disclosure, the step of acquiring EIS information for each of the plurality of states of charge may include a step of acquiring first EIS information, which is at least a part of the EIS information, resulting from the EIS being performed on the battery cell by a battery management device each time the state of charge of the battery cell increases at regular intervals during a charging process in which the battery cell is changed from a first state of charge to a second state of charge higher than the first state of charge; and a step of acquiring second EIS information, which is at least a part of the EIS information, resulting from the EIS being performed on the battery cell by the battery management device each time the state of charge of the battery cell decreases at regular intervals during a discharging process in which the battery cell is changed from the second state of charge to the first state of charge.

[0018] In addition, in one embodiment of the present disclosure, the step of generating the DRT information may include a battery analysis method including: calculating the DRT for first EIS information for the plurality of charge states during a charging process of the battery cell, and generating first DRT information for the plurality of charge states for the charging process, which is at least a portion of the DRT information; and calculating the DRT for second EIS information for the plurality of charge states during a discharging process of the battery cell, and generating second DRT information for the plurality of charge states for the discharging process, which is at least a portion of the DRT information.

[0019] In addition, in one embodiment of the present disclosure, a battery analysis method may be provided, further including a step of performing an analysis based on the characteristic image and estimating at least a portion of characteristic information of the battery cell, the characteristic information including at least a portion of material information, deterioration level information, and usage environment information.

[0020] In one embodiment of the present disclosure, the step of estimating at least a portion of the characteristic information of the battery cell may include a step of determining symmetry between a first characteristic image corresponding to a charging process of the battery cell and a second characteristic image corresponding to a discharging process of the battery cell, both included in the characteristic images; and a step of estimating at least a portion of the material information of the battery cell based on the symmetry.

[0021] In one embodiment of the present disclosure, the step of determining the symmetry may include a battery analysis method including the steps of: calculating a similarity between the first characteristic image and the second characteristic image; and determining that the symmetry does not exist between the first characteristic image and the second characteristic image based on the similarity being less than a threshold value.

[0022] In addition, in one embodiment of the present disclosure, the step of estimating at least a portion of the material information of the battery cell may include a step of determining that at least a portion of the negative electrode of the battery cell contains silicon oxide based on the first characteristic image and the second characteristic image in which the symmetry does not exist.

[0023] In addition, in one embodiment of the present disclosure, the step of estimating at least a portion of the material information of the battery cell may include a step of estimating a content ratio of the silicon oxide inversely proportional to a similarity between the first characteristic image and the second characteristic image corresponding to the symmetry.

[0024] In addition, in one embodiment of the present disclosure, the step of estimating at least a portion of the characteristic information of the battery cell may include a battery analysis method including: calculating the characteristic image and extracting a shape pattern corresponding to shading on the characteristic image; and estimating at least a portion of the usage environment information of the battery cell based on the shape pattern.

[0025] In addition, in one embodiment of the present disclosure, the step of estimating at least a portion of the characteristic information of the battery cell may include a battery analysis method including: calculating the characteristic image and obtaining shading amount information on the characteristic image; and estimating the deterioration level information of the battery cell based on the shading amount information.

[0026] Another aspect of the present disclosure may provide a server including: a memory for storing instructions; and a processor connected to the memory, wherein the processor is configured to: acquire electrochemical impedance spectroscopy (EIS) information for each of a plurality of states of charge determined by performing EIS on a battery cell at a plurality of different states of charge from a battery management device; generate distribution of relaxation times (DRT) information for the EIS information for each of the plurality of states of charge by calculating a distribution of relaxation times (DRT) for the EIS information for each of the plurality of states of charge; and generate a feature image corresponding to the battery cell, including information on impedance magnitude according to frequency calculated for each of the plurality of states of charge, based on the DRT information for each of the plurality of states of charge.

[0027] Yet another aspect of the present disclosure can provide a computer-readable non-transitory recording medium having recorded thereon a program for causing a server to execute the above-described battery analysis method.

[0028] Other specific details of the embodiments are included in the detailed description and drawings. [Effects of the Invention]

[0029] According to the proposed embodiment, one or more of the following advantages can be expected:

[0030] According to the embodiments of the present specification, a characteristic image can be generated based on the results of EIS and DRT for each state of charge of a battery cell.

[0031] Furthermore, according to the embodiments of the present specification, the characteristics of a target battery cell may be analyzed based on a characteristic image set including respective characteristic images for respective battery cells having respective characteristics.

[0032] Furthermore, according to the embodiments of the present specification, at least some of the material information, deterioration level information, and usage environment information of the target battery cell can be analyzed based on the characteristic image set.

[0033] Furthermore, according to the embodiments of the present specification, it is possible to non-destructively determine whether silicon oxide is present in the negative electrode of the battery cell by analyzing the characteristic image.

[0034] Furthermore, according to the embodiments of the present specification, it is also possible to analyze at least a part of the deterioration level information and the usage environment information of the battery cell based on the shape pattern or shading amount information of the characteristic image.

[0035] The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description of the claims. [Brief explanation of the drawings]

[0036] [Figure 1] 1 illustrates the relationship between a server that analyzes a battery according to an embodiment. [Figure 2] 1 is a flowchart of a battery analysis method according to an embodiment. [Figure 3a] An example of EIS information is shown in the figure. [Figure 3b] 3b illustrates an example of DRT information obtained by calculating the DRT for the EIS information of FIG. 3a. [Figure 4] 1 is a diagram illustrating an example of a characteristic image generated by a battery analysis method according to an embodiment; [Figure 5a] 10 is a diagram illustrating a relationship between characteristic images and DRT information according to an embodiment; [Figure 5b] 10 is a diagram illustrating a relationship between characteristic images and DRT information according to an embodiment; [Figure 6] 1 is an exemplary view showing a characteristic image corresponding to a battery cell including silicon oxide in an anode according to an embodiment; FIG. [Figure 7] FIG. 2 illustrates a block diagram of a server according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0037] The terms used in the embodiments are currently commonly used and general terms that have been selected as much as possible while taking into consideration the functions in the present disclosure, but these may change depending on the intentions or precedents of engineers in the field, the emergence of new technologies, etc. In addition, in certain cases, the applicant may arbitrarily select terms, and in such cases, the meanings thereof will be described in detail in the relevant description. Therefore, the terms used in the present disclosure are not simply defined by their names, but are defined based on the meanings of the terms and the overall content of the present disclosure.

[0038] Throughout the specification, when a part is said to "comprise" a certain element, this does not mean that it excludes other elements, but that it may further include other elements, unless otherwise specified.

[0039] Throughout the specification, the expression "at least one of a, b, and c" can encompass "a alone," "b alone," "c alone," "a and b," "a and c," "b and c," or "all of a, b, and c."

[0040] The "terminal" referred to below may be embodied as a computer or a portable terminal that can connect to a server or other terminals via a network. Here, the computer may include, for example, a notebook computer, desktop computer, or laptop computer equipped with a web browser, and the portable terminal may include, for example, a wireless communication device that ensures portability and mobility, such as communication terminals for International Mobile Telecommunication (IMT), Code Division Multiple Access (CDMA), W-CDMA, and Long Term Evolution (LTE), as well as all kinds of handheld wireless communication devices such as smartphones and tablet PCs.

[0041] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings so that those skilled in the art can easily implement the present disclosure. However, the present disclosure may be embodied in several different forms and is not limited to the embodiments described herein.

[0042] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0043] FIG. 1 illustrates the relationship between servers that analyze a battery according to one embodiment.

[0044] 1, a server 100 can operate in conjunction with a battery management device 200 that manages battery cells 300. Meanwhile, only components related to this embodiment are illustrated in FIG. 1. Therefore, a person skilled in the art of this embodiment can understand that other general components may be further included in addition to the components illustrated in FIG. 1.

[0045] The server 100 is a device that configures and provides various information. The server 100 may provide the configured information as a web page or an application screen, or may provide information in a format that can be displayed on a web page or an application screen on a device.

[0046] The battery management device 200 may include one or more sensors for measuring parameters such as the current, voltage, and temperature of the battery cells 300, and may include memory and a processor (not shown) for various operations. That is, the battery management device 200 operates based on the memory and processor, similar to the server 100, but may additionally include sensors to measure and calculate parameters of the battery cells 300. According to one embodiment, the battery management device 200 can perform electrochemical impedance spectroscopy (EIS) by applying an external AC power source, i.e., AC voltage or AC current, at various frequencies to the battery cells 300 and measuring the resulting current or voltage. Furthermore, such EIS can be performed at regular intervals between states of charge, at intervals between specified states of charge, or at any interval between states of charge while charging and discharging the battery. For example, if EIS is performed at regular intervals of the state of charge, EIS can be performed every time the SoC increases by 1% from 1% to 99% and then every time the SoC decreases by 1% from 99% to 1% while discharging again. In this way, the battery management unit 200 can perform EIS in conjunction with the charge / discharge process.

[0047] Here, the server 100 and the battery management unit 200 may be completely separate and independent entities, or may exist conceptually separated within a single device or system. That is, a single computing device having a control function for battery cells may perform all of the functions of the server 100 and the battery management unit 200 described below, and such an embodiment is also considered to fall within the scope of the present disclosure.

[0048] The battery cell 300 may be one of various secondary batteries. For example, it may be a conventionally widely used secondary battery such as a lithium-ion battery, or any one of various next-generation secondary batteries such as a lithium-sulfur battery, a lithium-metal battery, an all-solid-state battery, a metal-air battery, an aluminum-ion battery, a potassium-ion battery, or a zinc-air battery. According to an embodiment, as described below, the battery cell 300 may be a lithium-ion battery that may or may not contain silicon oxide in the anode. According to an embodiment, the battery analysis method of the present disclosure may be performed to determine whether the anode of a lithium-ion battery contains silicon oxide. Of course, the configuration in which the battery cell 300 is a lithium-ion battery and the analysis method of the present disclosure determines whether the anode contains silicon oxide is merely an example. That is, the battery cell 300 may be any type of secondary battery, and the analysis method of the present disclosure, described below, may be used to determine information about the material of the cathode or electrolyte, whether the anode contains components other than silicon oxide, the battery's usage environment, or its degradation level.

[0049] Hereinafter, a battery analysis method according to an embodiment of the present disclosure will be described with reference to FIG.

[0050] FIG. 2 is a flowchart of a battery analysis method according to one embodiment.

[0051] 2, in step S210, the server 100 may acquire EIS information for a plurality of states of charge confirmed by performing EIS on the battery cell 300 at a plurality of different states of charge. In step S220, the server 100 may generate DRT information for a plurality of states of charge by calculating a distribution of relaxation times (DRT) for the EIS information for the plurality of states of charge. In step S230, the server 100 may generate a feature image corresponding to the battery cell 300 including information on the magnitude of impedance according to frequency calculated for the plurality of states of charge based on the DRT information for the plurality of states of charge.

[0052] For ease of understanding of the present disclosure, the EIS and DRT information will be described below with reference to FIG.

[0053] 3a and 3b illustrate an example of EIS information and DRT information obtained by calculating the DRT. Referring to FIG. 3a, when EIS is performed on a battery cell, EIS information can be confirmed from a Nyquist plot, which shows the magnitude of impedance for each frequency, divided into real and imaginary axes. Furthermore, when DRT is calculated based on the Nyquist plot derived as in FIG. 3a, the magnitude of impedance for each frequency, which is log-scaled, can be derived as DRT information, as in FIG. 3b.

[0054] The EIS analysis or EIS measurement disclosed herein may refer to a process of performing electrochemical impedance spectroscopy on a battery cell, and the EIS information or EIS data may refer to information indicating impedance values ​​for a battery cell as a function of frequency. The EIS information may refer to impedance information that can be represented as a Nyquist plot. The DRT analysis disclosed herein may refer to an analytical technique for determining impedance values ​​for each frequency from the EIS information, and the DRT information may be a distribution graph that converts the impedance information into a distribution with respect to relaxation time. In other words, the DRT information may refer to a graph showing the distribution of impedance values ​​by frequency.

[0055] Each stage will be described in more detail below.

[0056] First, the server 100 can acquire EIS information for a plurality of states of charge of the battery cell 300. Here, such EIS information can be acquired from the battery management unit 200. As described above, the battery management unit 200 can measure the EIS information by performing EIS at a plurality of different states of charge while charging and discharging the battery cell 300. For example, the battery management unit 200 can perform EIS at regular intervals of states of charge, or can perform EIS at a plurality of specified states of charge. The server 100 can acquire such EIS information from the battery management unit 200.

[0057] According to one embodiment, the EIS information may include first EIS information, which is a result of EIS being performed on the battery cell 300 at a plurality of different charge states when the state of charge of the battery cell 300 increases during a charging process in which the battery cell 300 is changed from a first state of charge to a second state of charge, and second EIS information, which is a result of EIS being performed on the battery cell 300 at a plurality of different charge states when the state of charge of the battery cell 300 decreases during a discharging process in which the battery cell 300 is changed from the second state of charge to the first state of charge. That is, the battery management unit 200 measures the first EIS information during the charging process and the second EIS information during the discharging process, and the server 100 may acquire EIS information including the first and second EIS information from the battery management unit 200. As described above, during the charging and discharging processes, EIS may be performed every time the state of charge increases and decreases at regular intervals, or EIS may be performed for each of a plurality of pre-specified state of charge, or EIS may be performed for any desired state of charge.

[0058] Thereafter, the server 100 can generate DRT information for a plurality of states of charge by calculating the DRT for the EIS information acquired in this manner. At this time, the server 100 can generate first DRT information and second DRT information for the first EIS information and second EIS information corresponding to the charging process and the discharging process, respectively. Specifically, the server 100 can calculate the DRT for the first EIS information for a plurality of states of charge during the charging process of the battery cell 300, and generate first DRT information for a plurality of states of charge for the charging process, which is at least a part of the DRT information. Furthermore, the server 100 can calculate the DRT for the second EIS information for a plurality of states of charge during the discharging process of the battery cell 300, and generate second DRT information for a plurality of states of charge for the charging process, which is at least a part of the DRT information.

[0059] Thereafter, the server 100 may generate a feature image corresponding to the battery cell including frequency-dependent impedance magnitude information calculated for each of the plurality of states of charge based on the DRT information for each of the plurality of states of charge. Here, the feature image may be in the form of a two-dimensional image in which the first axis represents frequency, the second axis represents the state of charge, and the shading of each pixel is determined based on the impedance magnitude information calculated for each frequency and each of the plurality of states of charge.

[0060] Here, the impedance magnitude information may be represented as the degree of shading of each pixel as described above, but in one embodiment, it may be represented as height on a three-dimensional image. In other words, in this case, the feature image is a three-dimensional form, and can be generated by representing the value of a third axis perpendicular to the plane of a two-dimensional image, with the first axis representing frequency and the second axis representing charge state, as the impedance magnitude information.

[0061] According to one embodiment, the server 100 may calculate first DRT information corresponding to a charging process and second DRT information corresponding to a discharging process, respectively, and generate a first characteristic image and a second characteristic image. Specifically, the server 100 may check first magnitude information of impedance according to frequency for each of a plurality of states of charge for the charging process based on the first DRT information for each of a plurality of states of charge for the charging process of the battery cell 300, and generate the first characteristic image for the charging process, which is at least a part of the characteristic images, based on the first magnitude information. Furthermore, the server 100 may check second magnitude information of impedance according to frequency for each of a plurality of states of charge for the discharging process based on the second DRT information for each of a plurality of states of charge for the discharging process, and generate a second characteristic image for the discharging process, which is at least a part of the characteristic images, based on the second magnitude information.

[0062] According to one embodiment, the feature image may include a first feature image and a second feature image symmetrically arranged about a reference axis parallel to the first axis, i.e., the feature image may be generated such that the magnitude of the impedance represented by shading for the charging process and the discharging process for the same frequency can be easily identified.

[0063] As mentioned above, refer to FIG. 4 for an example of the generated feature images.

[0064] FIG. 4 is a diagram illustrating an example of a characteristic image generated by a battery analysis method according to an embodiment.

[0065] 4, the feature image may include a first feature image 110 and a second feature image 120 in a symmetrical structure about a reference axis 130. However, this structure is merely an example, and the feature image may include the first feature image 110 and the second feature image 120 in other ways, for example, symmetrical about a reference axis parallel to the second axis. Alternatively, the feature image may include the first feature image 110 and the second feature image 120 in a non-symmetrical structure, and all of these various embodiments of generating feature images are considered to be included in the present disclosure.

[0066] Meanwhile, the relationship between the first part 111 corresponding to the third charging state during the charging process and the second part 112 corresponding to the fourth charging state shown in FIG. 4 and the DRT information will be described with reference to FIG. 5a and FIG. 5.

[0067] 5a and 5b are diagrams illustrating a relationship between feature images and DRT information according to an embodiment.

[0068] First, Figure 5a may be a portion of DRT information corresponding to a third state of charge during a charging process. Comparing Figure 5a with the first portion 111 of Figure 3, it can be seen that, in the portion corresponding to the same frequency, the portion in Figure 5a where the magnitude of the impedance is large is represented by a darker shade in the first portion 111 of Figure 4. Similarly, Figure 5b may be a portion of DRT information corresponding to a fourth state of charge during a charging process. Comparing Figure 5b with the second portion 112 of Figure 3, it can be seen that, in the portion corresponding to the same frequency, the portion in Figure 5b where the magnitude of the impedance is large is represented by a darker shade in the second portion 112 of Figure 4.

[0069] According to one embodiment, as can be seen by comparing Figures 5a and 5b with the first and second parts 111 and 112 of Figure 4, the server 100 determines the degree of shading using frequency-specific impedance magnitude information indicated by DRT information for multiple charge states, thereby generating partial images for each charge state, and then stacking these in order according to the charge state value to generate a first characteristic image and a second characteristic image.

[0070] The generated characteristic image may include a unique pattern depending on the characteristics of each battery cell. For example, battery cells with different cathode or anode materials may have characteristic images with distinct patterns, and battery cells used in high-temperature and low-temperature environments may have different characteristic image patterns. Furthermore, battery cells with the same characteristics may have characteristic image patterns that are somewhat similar. The characteristics of a battery cell can be estimated based on characteristic images with such characteristics, and a first embodiment, which is one of the embodiments related to this, will be described below.

[0071] According to one embodiment, the server 100 may generate a characteristic image for the battery cell 300 as described above and then store the generated characteristic image in the database as at least a part of a characteristic image set. Alternatively, the server 100 may perform similar operations on a plurality of various reference battery cells, generate a characteristic image for each reference battery cell, and store the generated characteristic image in the database as at least a part of a characteristic image set.

[0072] According to one embodiment, each image included in the feature image set may be stored in association with at least a portion of reference feature information, including at least a portion of material information, degradation level information, and usage environment information of the corresponding reference battery cell. That is, the server 100 may calculate the DRT based on the EIS results of each reference battery cell categorized by the reference feature information, generate feature images using the DRT, and store the feature images in association with the reference feature information of each reference battery cell. For example, an administrator may prepare reference battery cells having respective characteristics, generate feature images for them through the server 100, and build a feature image set.

[0073] In the above description, examples have been given in which the reference characteristic information includes material information, deterioration level information, and usage environment information, but those skilled in the art will understand that the reference characteristic information is not limited to these and may include various characteristics related to the battery cell.

[0074] According to one embodiment, after the feature image set is constructed, the server 100 can estimate feature information of a target battery cell based on the feature image set. That is, similar to the above process, when multiple pieces of target EIS information are generated by performing EIS on the target battery cell at multiple different charge states during charging and discharging processes by the battery management device, the server 100 can acquire the target EIS information and calculate the DRT to calculate target DRT information. Thereafter, the server 100 can generate a target image for the target battery cell based on the target DRT information. The server 100 can estimate feature information of the target battery cell based on the target image and feature image set thus generated.

[0075] Specifically, the server 100 may calculate the similarity between the target image and the images included in the feature image set. Thereafter, reference feature information associated with the image with the highest similarity may be estimated as feature information of the target battery cell. As described above, since a battery cell has its own unique feature image according to its characteristics, the target battery cell may be estimated to have similar features to the reference battery cell corresponding to the image with the highest similarity among the images included in the feature image set.

[0076] As a variation of the embodiment for estimating feature information of a target battery cell based on the similarity to an image included in a feature image set, an embodiment using a neural network will be described. That is, according to one embodiment, the server 100 is equipped with a neural network trained based on the feature image set and can estimate feature information of a target battery cell using the neural network. The neural network can be trained using the feature image set and reference feature information associated with each image included therein as training data and ground truth information related thereto. The neural network can be implemented using a widely known convolutional neural network (CNN) or various methods used for image processing. When the neural network is implemented using a CNN, the neural network can operate on each image included in the feature image set as training data to generate training output information, compare the training output information with the reference feature information as ground truth to generate a loss, and then back-propagate the loss to learn at least some of its parameters. The neural network trained in this manner can operate on a target image and output feature information estimated to be possessed by the target battery cell.

[0077] Next, a second embodiment for estimating the characteristics of a battery cell based on a characteristic image will be described.

[0078] According to one embodiment, the server 100 performs an analysis based on the characteristic image to estimate at least a portion of the characteristic information of the battery cell 300. The characteristic information may include at least a portion of the material information, deterioration level information, and usage environment information of the battery cell 300. Examples of estimating the material information, deterioration level information, and usage environment information will be described below.

[0079] First, the server 100 may determine symmetry, based on the maximum SoC, between a first feature image corresponding to the charging process of the battery cell 300 and a second feature image corresponding to the discharging process of the battery cell 300, included in the feature image. Here, the determination of symmetry may be performed based on the similarity when the first feature image and the second feature image are compared on the same scale in terms of state of charge and frequency. That is, the first axis of the first feature image and the second feature image may be set to frequency on a logarithmic scale, the second axis to state of charge, and the first point on the second axis may be set to an SoC of 0% and the second point to an SoC of 100%, respectively, and then the similarity between the first feature image and the second feature image may be calculated after the same scaling. If the similarity is equal to or greater than a threshold value, the server 100 may determine that there is symmetry between the first feature image and the second feature image constituting the feature image. If the similarity is less than the threshold value, the server 100 may determine that there is no symmetry between the feature images. Here, the similarity may correspond to the similarity between the first and second characteristic images during the charge / discharge process of a lithium-ion battery that does not contain any silicon oxide in the anode, and the similarity between such images may be calculated using various algorithms known in the art. For example, the shading values ​​of the respective coordinates of the first and second characteristic images may be compared to calculate the difference values ​​for each coordinate, and the similarity may be calculated as being inversely proportional to the sum of the difference values ​​for all coordinates.

[0080] In this case, if symmetry exists, the server 100 can determine that the battery cell 300 does not contain silicon oxide in the negative electrode. When silicon oxide is contained in the negative electrode of a lithium-ion battery, hysteresis occurs in the electrode state during the charge and discharge process due to a characteristic reaction of silicon oxide in the negative electrode. Therefore, when silicon oxide is contained in the negative electrode, the first and second characteristic images for the charge and discharge processes may be different, whereas when silicon oxide is not contained in the negative electrode, the first and second characteristic images may be the same because hysteresis does not occur. Using these characteristics, it is possible to determine whether the negative electrode contains silicon oxide based on the characteristic images. To consider this embodiment, reference is now made to FIGS. 4 and 6, which were previously described.

[0081] Referring back to FIG. 4 , it can be seen that the first feature image 110 and the second feature image 120 are symmetrical about the reference axis 130. The similarity calculation described above can be performed after flipping either the first feature image 110 or the second feature image 120 upside down about the reference axis 130. When either one of the first feature image 110 and the second feature image 120 is flipped upside down about the reference axis 130, the shading of corresponding coordinates in the first feature image 110 and the second feature image 120 indicates the same impedance magnitude for the same state of charge and frequency. Therefore, as described above, the first feature image 110 and the second feature image 120 can be considered to be scaled to the same scale and the similarity calculation can be performed. When the first feature image 110 and the second feature image 120 are symmetrical as shown in FIG. 4 , the server 100 can determine that the battery cell 300 does not contain silicon oxide in the negative electrode. Also, with reference to FIG. 4, the symmetry referred to in this disclosure can also be understood as the symmetry of first feature image 110 and second feature image 120 about reference axis 130.

[0082] FIG. 6 is an example of a characteristic image corresponding to a battery cell including silicon oxide in the negative electrode according to one embodiment.

[0083] 6 is an example diagram of a characteristic image corresponding to a battery cell including silicon oxide in the negative electrode, and it can be seen that the first characteristic image 110 and the second characteristic image 120 are asymmetric about the reference axis 130. As described above, when silicon oxide is included in the negative electrode, the charge and discharge process is asymmetric due to hysteresis, and therefore the characteristic images, such as the first characteristic image 110 and the second characteristic image 120, can be derived asymmetrically about the reference axis 130. If it is determined that the first and second characteristic images 110 and 120 are asymmetric as shown in FIG. 6, the server 100 can determine that the battery cell 300 includes silicon oxide in the negative electrode.

[0084] According to one embodiment, server 100 can estimate the proportion of silicon oxide contained in the negative electrode based on the similarity. The more silicon oxide contained in the negative electrode, the greater the hysteresis loop, and the more asymmetric the charge and discharge processes may be. The more asymmetric the charge and discharge processes, the lower the calculated similarity between the first and second characteristic images. Therefore, server 100 can estimate the silicon oxide content ratio inversely proportional to the similarity.

[0085] The above-described embodiment may be applied to cases where the negative electrode contains not only silicon oxide but also other silicon-containing compounds, such as silicon carbide, silicon nitride, or silicon alloys, or where a component that induces hysteresis is contained in at least a portion of the positive electrode, negative electrode, or electrolyte.

[0086] Meanwhile, according to one embodiment, the server 100 may calculate the characteristic image and extract a shape pattern corresponding to the shading on the characteristic image. The server 100 may then estimate at least a portion of the usage environment information of the battery cell 300 based on the shape pattern. For example, a battery cell continuously used in a low-temperature or high-temperature environment may exhibit a characteristic shape pattern in the characteristic image, such as a straight line or an upward or downward curve at a specific frequency. A characteristic shape pattern may also be present when rapid charging is frequently performed or when a battery is installed in a vehicle and charging and discharging are repeated at short intervals due to regenerative braking, etc. Based on such a shape pattern, the server 100 may estimate at least a portion of the usage environment information of the battery cell 300. Additionally, according to one embodiment, the server 100 may also estimate at least a portion of the material information of the battery cell 300 based on the shape pattern. For example, if the shape pattern appearing on the characteristic image is similar to a shape pattern that appears characteristically for a specific material contained in the positive electrode or another specific material contained in the electrolyte, the server 100 may infer that the specific material is contained in the positive electrode or the electrolyte.

[0087] According to one embodiment, the server 100 may calculate the feature image to obtain shading amount information on the feature image. The server 100 may then estimate deterioration level information of the battery cell 300 based on the shading amount information. When the battery cell 300 has been used for a long time and is highly deteriorated, its internal resistance increases. As the internal resistance increases, the magnitude of the impedance increases, resulting in darker shading at each coordinate in the feature image. Therefore, for a battery cell 300 that has been used for a long time, the overall amount of shading on the feature image is likely to be greater than for a battery cell that has not been used in a long time. Therefore, the server 100 may estimate deterioration level information based on the shading amount information such that the amount of shading is proportional to the degree of deterioration. Through the above method, feature information of a target battery cell can be accurately estimated nondestructively.

[0088] FIG. 7 shows a block diagram of a server according to one embodiment.

[0089] According to one embodiment, the server 100 may include a memory 101 and a processor 102. The server 100 illustrated in FIG. 7 only shows components relevant to this embodiment. Therefore, a person skilled in the art will understand that general components may be included in addition to the components illustrated in FIG. 7. In one embodiment, the processor 102 may be included in a controller.

[0090] The processor 102 can control the overall operation of the server 100 and process data and signals. The processor 102 can be configured with at least one hardware unit. The processor 102 can also be operated by one or more software modules generated by executing program code stored in the memory 101. Because the processor 102 can include memory, the processor 102 can execute the program code stored in the memory to control the overall operation of the server 100 and process data and signals.

[0091] The processor 102 may be configured to: acquire electrochemical impedance spectroscopy (EIS) information for each of a plurality of states of charge, which is confirmed by performing EIS on the battery cell at a plurality of different states of charge, from the battery management device; generate distribution of relaxation times (DRT) information for the EIS information for each of the plurality of states of charge by calculating DRT information for the plurality of states of charge; and generate a feature image corresponding to the battery cell, including information on the magnitude of impedance according to frequency, calculated for each of the plurality of states of charge, based on the DRT information for the plurality of states of charge.

[0092] Depending on the embodiment, the server 100 may further include a transceiver for wired / wireless communication. The server 100 may communicate with an external electronic device (e.g., the battery management device 200) using the transceiver. The external electronic device may be a terminal or a server. In addition, communication technologies used by the transceiver may include Global System for Mobile communication (GSM), Code Division Multi Access (CDMA), Long Term Evolution (LTE), 5G, Wireless LAN (WLAN), Wireless Fidelity (Wi-Fi), Bluetooth (registered trademark), Radio Frequency Identification (RFID), Infrared Data Association (IrDA), ZigBee, Near Field Communication (NFC), etc.

[0093] The server according to the above-described embodiments may include a processor, memory for storing and executing program data, permanent storage such as a disk drive, a communication port for communicating with external devices, and user interface devices such as a touch panel, keys, buttons, etc. Methods embodied as software modules or algorithms may be stored on a computer-readable recording medium as computer-readable code or program instructions executable on the processor. Examples of computer-readable recording media include magnetic recording media (e.g., read-only memory (ROM), random-access memory (RAM), floppy disks, hard disks, etc.) and optically readable media (e.g., CD-ROMs, Digital Versatile Discs (DVDs)). The computer-readable recording media may be distributed across computer systems connected via a network, allowing the computer-readable code to be stored and executed in a distributed manner. The medium is computer-readable, can be stored in memory, and can be executed by the processor.

[0094] The present embodiments may be illustrated with functional block configurations and various processing steps. These functional blocks may be embodied in various hardware and / or software components that perform specific functions. For example, the embodiments may employ integrated circuit configurations such as memory, processing, logic, look-up tables, etc., that can perform various functions under the control of one or more microprocessors or other control devices. Just as components may be implemented in software programming or software elements, the present embodiments include various algorithms embodied in a combination of data structures, processes, routines, or other programming components, and may be embodied in programming or scripting languages ​​such as C, C++, Java, assembler, etc. Functional aspects may be embodied in algorithms executed on one or more processors. The present embodiments may also employ conventional techniques for electronic configuration, signal processing, and / or data processing. Terms such as "mechanism," "element," "means," and "component" may be used broadly and are not limited to mechanical and physical components. These terms may include the meaning of a series of software routines in cooperation with a processor or the like.

[0095] The above-described embodiments are by way of example only, and other embodiments may be embodied within the scope of the following claims.

Claims

1. In the battery analysis method performed on the server, performing electrochemical impedance spectroscopy (EIS) on the battery cell at different states of charge, and acquiring EIS information for each of the different states of charge; generating distribution of relaxation times (DRTs) for the plurality of state-of-charge EIS information by calculating DRTs; generating a characteristic image corresponding to the battery cell, the characteristic image including frequency-dependent impedance magnitude information calculated for each of the plurality of charge states based on the DRT information for each of the plurality of charge states.

2. 2. The battery analysis method of claim 1, wherein generating the characteristic image comprises generating the characteristic image in the form of a two-dimensional image, in which a first axis is the frequency, a second axis is the state of charge, and a shading level of each pixel is determined based on the magnitude information of the impedance calculated for each of the frequencies and the plurality of states of charge, based on the DRT information for each of the plurality of states of charge.

3. The step of generating the feature image comprises: determining first magnitude information of impedance according to the frequency for each of the plurality of charge states for the charging process based on first DRT information for each of the plurality of charge states in the charging process of the battery cell, and generating a first characteristic image for the charging process, which is at least a part of the characteristic images, based on the first magnitude information; 3. The battery analysis method of claim 2, further comprising: determining second magnitude information of impedance according to the frequency for each of the plurality of states of charge for the discharging process based on second DRT information for each of the plurality of states of charge during the discharging process of the battery cell; and generating a second characteristic image for the discharging process, which is at least a part of the characteristic image, based on the second magnitude information.

4. The battery analysis method of claim 3 , wherein the feature image includes the first feature image and the second feature image symmetrically about a reference axis parallel to the first axis.

5. The battery analysis method of claim 1 , further comprising: storing a set of characteristic images for a plurality of reference battery cells, including the characteristic image, in a database.

6. 6. The battery analysis method of claim 5, wherein the characteristic image set is stored in association with at least a portion of reference characteristic information, including at least a portion of material information, deterioration level information, and usage environment information of each reference battery cell corresponding to each image included in the characteristic image set.

7. performing EIS at a plurality of different charge states of a target battery cell to obtain a plurality of target EIS information, calculating DRT for the target EIS information to obtain target DRT information, and generating a target image for the target battery cell based on the target DRT information; The battery analysis method of claim 5 , further comprising: estimating feature information of the target battery cell based on the feature image set and the target image.

8. 10. The battery analysis method of claim 7, wherein estimating the feature information of the target battery cell comprises calculating a similarity between the target image and an image included in the feature image set, and then estimating reference feature information associated with an image having the highest similarity as the feature information of the target battery cell.

9. 10. The battery analysis method of claim 7, wherein the step of estimating the feature information of the target battery cell comprises inputting the target image to a neural network trained by inputting the feature image set and reference feature information corresponding to each image included in the feature image set as learning data and corresponding ground truth information, and processing the target image through the neural network to estimate output feature information as the feature information of the target battery cell.

10. The step of acquiring EIS information for each of the plurality of charging states includes: In a charging process of changing the battery cell from a first state of charge to a second state of charge higher than the first state of charge, acquiring first EIS information, which is at least a part of the EIS information, the first EIS information being a result of EIS performed on the battery cell by a battery management device, every time the state of charge of the battery cell increases at a certain interval; and acquiring second EIS information, which is at least a part of the EIS information, obtained by performing EIS on the battery cell by the battery management device every time the state of charge of the battery cell decreases at regular intervals during a discharging process that changes the battery cell from the second state of charge to the first state of charge.

11. The step of generating the DRT information includes: Calculating DRT for first EIS information for each of the plurality of charge states in a charging process of the battery cell, and generating first DRT information for each of the plurality of charge states for the charging process, which is at least a portion of the DRT information; 2. The battery analysis method of claim 1, further comprising: calculating a DRT for the second EIS information for each of the plurality of states of charge during the discharging process of the battery cell; and generating second DRT information for each of the plurality of states of charge for the discharging process, which is at least a portion of the DRT information.

12. 2. The battery analysis method of claim 1, further comprising: performing an analysis based on the characteristic image to estimate at least a portion of characteristic information of the battery cell, the characteristic information including at least a portion of material information, deterioration level information, and usage environment information.

13. The step of estimating at least a portion of the characteristic information of the battery cell includes: determining symmetry between a first feature image corresponding to a charging process of the battery cell and a second feature image corresponding to a discharging process of the battery cell, the first feature image and the second feature image being included in the feature images; and estimating at least a portion of the material information of the battery cell based on the symmetry.

14. The step of determining symmetry comprises: calculating a similarity between the first feature image and the second feature image; 14. The battery analysis method of claim 13, further comprising determining that the symmetry does not exist between the first characteristic image and the second characteristic image based on the similarity being less than a threshold value.

15. 14. The battery analysis method of claim 13, wherein the step of estimating at least a portion of the material information of the battery cell includes a step of determining that at least a portion of the negative electrode of the battery cell contains silicon oxide based on the first characteristic image and the second characteristic image in which the symmetry does not exist.

16. 16. The battery analysis method of claim 15, wherein estimating at least a portion of the material information of the battery cell includes estimating the silicon oxide content ratio in inverse proportion to a similarity between the first characteristic image and the second characteristic image corresponding to the symmetry.

17. The step of estimating at least a portion of the characteristic information of the battery cell includes: calculating the characteristic image and extracting a shape pattern corresponding to the shading on the characteristic image; The battery analysis method according to claim 12 , further comprising: estimating at least a portion of the usage environment information of the battery cell based on the shape pattern.

18. The step of estimating at least a portion of the characteristic information of the battery cell includes: calculating the characteristic image and obtaining shading information on the characteristic image; The battery analysis method according to claim 12 , further comprising: estimating the deterioration level information of the battery cell based on the shading amount information.

19. A computer-readable non-transitory recording medium having recorded thereon a program for causing a server to execute the battery analysis method according to any one of claims 1 to 18.

20. a server, a memory for storing instruction words; a processor coupled to the memory; The processor: performing electrochemical impedance spectroscopy (EIS) on the battery cell at a plurality of different states of charge, and acquiring EIS information for each of the plurality of states of charge from a battery management device; generating DRT information for each of the plurality of states of charge by calculating a distribution of relaxation times (DRT) for the EIS information for each of the plurality of states of charge; The server is configured to generate a characteristic image corresponding to the battery cell, the characteristic image including frequency-dependent impedance magnitude information calculated for each of the plurality of charge states, based on the DRT information for each of the plurality of charge states.

Images