Battery cell process data analysis system and method

Abstract

A battery cell process data analysis system according to one embodiment includes a process data acquisition unit that acquires process data for a battery cell, a predicted capacity estimation unit that estimates a predicted capacity of the battery cell based on the process data, an actual capacity measurement unit that measures an actual capacity of the battery cell, and a controller that calculates a difference between the predicted capacity of the battery cell and the actual capacity of the battery cell, and determines process factors that affect the difference in capacity using a trained artificial intelligence model.

Description

[Technical field]

[0001] The present invention claims the benefit of priority based on Korean Patent Application No. 10-2022-0037600, filed March 25, 2022, and all contents disclosed in the documents of the Korean patent application are incorporated by reference into this specification.

[0002] The embodiments disclosed in this document relate to a battery cell process data analysis system and a battery cell process data analysis method using the same. [Background technology]

[0003] Lithium polymer batteries, a type of secondary battery, can be repeatedly charged and discharged by reversible mutual conversion between chemical energy and electrical energy. They have the advantages of long life and large capacity, and are used in a variety of applications, such as batteries for portable electronic devices and electric vehicles. Battery capacity, one of the most important performance indicators of a battery cell, is affected by numerous process factors involved in the manufacturing process. Conventionally, the process was managed by analyzing the simple correlation with capacity, centered on the design factors of the battery cell (i.e., variables that can be controlled in the battery design process). However, the actual manufacturing process of a battery cell consists of multiple discontinuous processes with different characteristics, such as an electrode process, an assembly process, and an activation process, and various process factors other than the design factors may affect the capacity of the battery cell. The effects of such process factors on the capacity of a battery cell are generally irregular, and multiple processes affect the capacity simultaneously, not a single process, making it difficult to clearly grasp these correlations. Summary of the Invention [Problem to be solved by the invention]

[0004] One objective of the embodiments disclosed in this document is to provide a battery cell process data analysis system and method, which can find out the process factors that affect the capacity of the battery cell through artificial intelligence-based data analysis.

[0005] Another objective of the embodiments disclosed herein is to improve the capacity quality of produced battery cells by finding process factors that have a positive impact on the capacity of the battery cells (e.g., making the actual capacity larger than the design capacity) and a negative impact on the capacity of the battery cells (e.g., making the actual capacity smaller than the design capacity) and managing the process factors.

[0006] The technical problems of the embodiments disclosed in this document are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0007] In order to achieve the above-mentioned object, one embodiment of a battery cell process data analysis system includes a process data acquisition unit that acquires process data of a battery cell, a predicted capacity estimation unit that estimates a predicted capacity of the battery cell based on the process data, an actual capacity measurement unit that measures an actual capacity of the battery cell, and a controller that calculates a difference between the predicted capacity of the battery cell and the actual capacity of the battery cell, and determines process factors that affect the difference in capacity using a trained artificial intelligence model.

[0008] According to one embodiment, the predicted capacity estimation unit may calculate the size of an electrode of each mono cell based on the measured values ​​of the individual mono cells, and estimate the predicted capacity of a battery cell in which a plurality of mono cells are stacked based on the size of the electrode and the amount of electrode active material loaded on the electrode.

[0009] According to one embodiment, the controller acquires distribution data of actual capacity relative to predicted capacity for a plurality of battery cells manufactured through a process, classifies the distribution data into two or more groups according to the magnitude of the actual capacity of the battery cell relative to the predicted capacity using a first artificial intelligence model, and analyzes the distribution data belonging to each group and the process data of the corresponding battery cell using a second artificial intelligence model, thereby calculating the importance indicating the effect of the process factors on the capacity of the battery cell.

[0010] According to one embodiment, the second artificial intelligence model can be trained to calculate the importance of each process factor based on changes in the distribution data depending on the presence or absence of two or more process factors.

[0011] According to one embodiment, the controller classifies the distribution data into a first group when the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or greater than a first threshold, and classifies the distribution data into a second group when the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or less than a second threshold. For distribution data belonging to the first group, process factors that affect the difference between the predicted capacity and the actual capacity of the battery cell are determined as positive process factors, and for distribution data belonging to the second group, process factors that affect the difference between the predicted capacity and the actual capacity of the battery cell are determined as negative process factors.

[0012] According to one embodiment, the process data of the battery cell includes data associated with at least one of an electrode process, an assembly process, and an activation process, the process data acquisition unit generates integrated process data by linking each process data associated with different processes using a key index, and the controller can use the integrated process data to determine a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell, among process factors associated with different processes.

[0013] In order to achieve the above-mentioned object, one embodiment of a method for analyzing process data of a battery cell includes the steps of acquiring process data of a battery cell, estimating a predicted capacity of the battery cell based on the process data, measuring an actual capacity of the battery cell, calculating a difference between the actual capacity and the predicted capacity of the battery cell, and determining process factors that affect the difference in capacity using a trained artificial intelligence model.

[0014] According to one embodiment, the step of estimating the predicted capacity may include a step of calculating a size of an electrode of the mono cell based on a measurement value of the individual mono cell, and a step of estimating a predicted capacity of a battery cell in which a plurality of individual mono cells are stacked based on the size of the electrode and the amount of electrode active material loaded on the electrode.

[0015] According to one embodiment, the step of determining the process factors includes the steps of: acquiring distribution data of actual capacity versus predicted capacity for a plurality of battery cells manufactured through a process; classifying the distribution data into two or more groups according to the magnitude of the actual capacity of the battery cells versus the predicted capacity using a first artificial intelligence model; and calculating the importance indicating the influence of the process factors on the capacity of the battery cells by analyzing the distribution data belonging to each group and the process data of the corresponding battery cells using a second artificial intelligence model.

[0016] According to one embodiment, the second artificial intelligence model can be trained to calculate the importance of each process factor based on changes in the distribution data due to the presence or absence of two or more pre-set process factors.

[0017] According to one embodiment, in the step of classifying the distribution data into two or more groups, if the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or greater than a first threshold, the distribution data is classified into a first group, and if the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or less than a second threshold, the distribution data is classified into a second group. In the step of determining the process factors, for the distribution data belonging to the first group, a process factor that affects the difference between the predicted capacity and the actual capacity of the battery cell can be determined as a positive process factor, and for the distribution data belonging to the second group, a process factor that affects the difference between the predicted capacity and the actual capacity of the battery cell can be determined as a negative process factor.

[0018] According to one embodiment, the process data of the battery cell includes data associated with at least one of an electrode process, an assembly process, and an activation process. In the step of acquiring the process data, integrated process data is generated by linking each of the process data associated with different processes using a key index. In the step of determining the process factors, the integrated process data can be used to determine a process factor that affects the difference between the predicted capacity and the actual capacity of the battery cell, among the process factors associated with the different processes. Effect of the Invention

[0019] According to the battery cell process data analysis system and method of the above embodiment, process factors that affect the capacity of a battery cell can be found by artificial intelligence-based data analysis. In addition, by linking data of different processes using a key index, it is possible to integrate and manage a plurality of process factors that affect the battery capacity at each process step.

[0020] Furthermore, by finding process factors that have a positive effect on the capacity of the battery cell (e.g., making the actual capacity larger than the predicted capacity) and process factors that have a negative effect (e.g., making the actual capacity smaller than the predicted capacity), and by controlling the process factors, the capacity quality of the produced battery cells can be improved. In addition, various other benefits may be provided by this document, either directly or indirectly.

[0021] In order to more clearly describe the embodiments disclosed in this document or the technical solutions of the prior art, drawings necessary for the description of the embodiments are briefly introduced below. It should be understood that the following drawings are only for describing the embodiments of the present specification and are not intended to be limiting. In addition, for the sake of clarity, the expressions of some components in the drawings may be exaggerated or omitted. [Brief description of the drawings]

[0022] [Figure 1] 1 is a graph showing predicted and actual distributions of measured actual capacity versus calculated predicted capacity for a number of battery cells. [Diagram 2] 1 is a block diagram showing a configuration of a battery cell process data analysis system according to an embodiment; [Diagram 3] 1 is a schematic diagram illustrating how process factors in multiple process steps involved in the manufacturing process of a battery cell affect the capacity of the completed battery cell. [Figure 4] 10 is a schematic diagram showing how process data for a plurality of processes included in a battery cell manufacturing process is linked using a key index. [Diagram 5] This shows how the electrodes are sized to calculate the predicted capacity of a battery monocell. [Figure 6] 1 is a graph showing distribution data of measured actual capacity versus predicted capacity for multiple battery cells. [Figure 7] 1 illustrates capacity distribution data of battery cells classified into two groups using an artificial intelligence model according to an embodiment. [Figure 8] 1 illustrates capacity distribution data of battery cells classified into three groups using an artificial intelligence model according to an embodiment. [Figure 9] 4 is a flowchart illustrating a method for analyzing process data of a battery cell according to an embodiment. [Figure 10] 10 is a flowchart showing in more detail the steps of determining process parameters in FIG. 9. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Hereinafter, the embodiments disclosed in this document will be described in detail with reference to the exemplary drawings. When referring to components in each drawing, it should be noted that the same components are referred to by the same reference numerals as much as possible when they are displayed in other drawings. In addition, when describing the embodiments disclosed in this document, if a detailed description of related known configurations or functions is deemed to hinder understanding of the embodiments disclosed in this document, the detailed description will be omitted.

[0024] The terms used in this document are currently selected as widely used as possible while taking into consideration the functions, but these may vary depending on the intentions or practices of the technical personnel in the field, or the emergence of new technologies. In addition, in certain cases, the applicant may arbitrarily select terms, and in such cases, the meanings of these terms are described in the description of the specification. Therefore, it is clear that the terms used in this document must be interpreted based on the substantial meanings of the terms and the overall content of this document, rather than simply the names of the terms. In addition, the terms used in this document are merely used to describe certain embodiments, and are not intended to limit the scope of other embodiments.

[0025] In this document, singular expressions may include plural expressions unless the context clearly indicates otherwise. In addition, in this document, expressions such as "first" and "second" are used to distinguish components from each other and do not indicate the order or sequence between the components.

[0026] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of a battery cell process data analysis system and a battery cell process data analysis method using the same will be described with reference to the drawings. FIG. 1 is a graph showing the distribution of predicted and actual capacities measured for multiple battery cells. As shown in the figure, according to the predicted distribution, the predicted and actual capacities have a linear relationship, but in reality, they tend to be distributed more widely due to the influence of various process factors. That is, in the actual distribution, due to the influence of unexpected process factors, the measured actual capacity may be smaller (lower part of the predicted distribution range) or larger (upper part of the predicted distribution range) than the calculated predicted capacity. The influence of such process factors on the capacity of a battery cell is generally irregular, and multiple processes affect the capacity simultaneously rather than a single process, making it difficult to clearly grasp the correlation between them.

[0027] FIG. 2 is a block diagram showing the configuration of a battery cell process data analysis system according to an embodiment for solving the above problem. 2, the battery cell process data analysis system 10 according to the embodiment may include a process data acquisition unit 110, a predicted capacity estimation unit 120, an actual capacity measurement unit 130, and a controller 140. The components shown in the block diagram are divided according to their respective functions and roles, and each block does not necessarily have to be realized by independent hardware or software. For example, the divided components may actually be realized by one device or program, or one component may be realized as a combination of multiple devices and programs.

[0028] The process data acquisition unit 110 acquires process data of the battery cell. Here, the process data of the battery cell is data related to a plurality of processes included in the battery manufacturing process, such as an electrode process, an assembly process, and an activation process, and may include data of a plurality of process factors (i.e., process variables) that affect the capacity of the produced battery cell. For example, in the electrode process, the loading amount of electrode material per unit area, the size and thickness of the electrode, etc. correspond to the process factors, in the assembly process, the loading method of the positive electrode, separator, and negative electrode, the amount of electrolyte, etc. correspond to the process factors, and in the activation process, the number of times the cell is charged and discharged, the temperature and pressure during charging and discharging, etc. correspond to the process factors. The types of data listed above are merely examples, and various process data related to process steps that may affect the capacity of the battery cell may be used.

[0029] FIG. 3 is a schematic diagram showing how process factors in multiple process steps involved in the manufacturing process of a battery cell affect the capacity of the completed battery cell. 3, various processes (first process to third process) and process factors (process factor 1 to process factor 6) involved in these processes have an effect on determining the capacity of a battery cell. In order to find out, among these process factors, the process factor that has a large effect (or a process factor that has a positive / negative effect) on the difference between the theoretically calculated predicted capacity and the actual measured capacity, the process data acquisition unit 110 can collect process data for each process.

[0030] According to an embodiment, the process data acquisition unit 110 can generate integrated process data by linking each process data related to different processes using a key index. Conventionally, a skilled engineer directly finds and manages design factors or process factors that affect the capacity of a battery cell based on his / her personal experience. Although this method is effective for managing individual processes (e.g., one of an electrode process, an assembly process, and an activation process), it is difficult to grasp the relationship between process factors of multiple process steps, and therefore it is not effective as the process steps become more complex. That is, since some process factors may affect each other as dependent variables rather than independent variables for battery capacity, it is advantageous to integrate and manage the process data of multiple process steps. For this purpose, it is necessary to link these data by considering the relationship between the process data of each process.

[0031] FIG. 4 is a schematic diagram showing how process data for a plurality of processes included in a battery cell manufacturing process is linked using a key index. 4, for a plurality of battery cells (AAA to ZZZ), process data (electrode 1, electrode 2, ..., electrode n) of a first process (e.g., electrode process), process data (assembly 1, assembly 2, ..., assembly n) of a second process (e.g., assembly process), and process data (activation 1, activation 2, ..., activation n) of a third process (e.g., activation process) for each of the battery cells are linked using key indexes to generate integrated process data. In this case, the key indexes may be set in consideration of the relationship between the processes.

[0032] The key index may vary depending on the type of data generated in the process. For example, in the case of a mono cell assembly process, an ID is assigned to each mono cell, so a key index can be set that links the IDs of multiple mono cells to the ID of one completed cell. In this way, in the case of an assembly process, each piece of data exists discontinuously, so it is possible to link processes by tracking the ID.

[0033] In the case of the electrode process, continuous time series data exists under one electrode ID due to the roll-to-roll process. Therefore, when linking the electrode process and the assembly process, the process data can be linked together by tracking the ID of the electrode input into the assembly process and the data in chronological or reverse chronological order within one electrode ID.

[0034] The integrated process data is input to an artificial intelligence model, which will be described later, and the artificial intelligence model can determine process factors that have a significant effect on the capacity of the battery cell based on the integrated process data. By integrating and managing the process data of multiple process steps in this manner, the effect of each process factor on the battery capacity can be easily and accurately understood.

[0035] The predicted capacity estimating unit 120 can estimate a predicted capacity of the battery cell based on the process data. Here, the predicted capacity means a theoretical maximum capacity calculated based on the design variables of the battery cell. According to an embodiment, the size of the electrodes (e.g., the size of the positive or negative electrode) can be calculated from vision data obtained by photographing the upper and lower parts of the surface of a monocell (a unit cell composed of a positive electrode, a negative electrode, and a separator), and the theoretical capacity that the battery cell can electrically have can be estimated based on the size of the electrodes and the amount of electrode active material loaded on the electrodes.

[0036] FIG. 5 shows how the electrodes are sized to calculate the expected capacity of a battery monocell. Referring to FIG. 5, the monocell 20 is composed of a positive electrode 210, a negative electrode 220, and a separator 230 inserted between the positive electrode and the negative electrode to separate them. Using vision data obtained by photographing the upper and lower parts of the monocell surface, the distance from each part of the positive electrode 210 to the end of the separator 230 (see the arrows in FIG. 5) can be known. Since the lateral length (W) of the separator 230 is already known as a design variable, the lateral length of the positive electrode 210 can be calculated by subtracting the lateral distance (see the arrows) between the positive electrode 210 and the separator 230 from the lateral length (W) of the separator 230. Similarly, since the longitudinal length (H) of the separator 230 is also already known as a design variable, the longitudinal length of the positive electrode 210 can be calculated by subtracting the longitudinal distance (see the arrows) between the positive electrode 210 and the separator 230 from the longitudinal length (H) of the separator 230.

[0037] The size of the positive electrode can be calculated from the horizontal and vertical lengths of the positive electrode 210 calculated in this manner, and the theoretical predicted capacity of each mono cell 20 can be calculated by substituting the amount of active material per unit area. Furthermore, since one battery cell has a structure in which a plurality of mono cells are stacked, it is possible to estimate the theoretical predicted capacity of the battery cell by adding up all the predicted capacities of each mono cell. The above-mentioned method of calculating the size of the electrode assumes that a specified design variable is taken into account, but this is merely an example. According to an embodiment, a method of calculating the size of the electrode can be considered by identifying a specified individual (e.g., positive electrode, negative electrode, separator) using an image processing algorithm in vision data obtained by photographing the upper and lower parts of the mono cell surface, and calculating the size of the identified individual.

[0038] Generally, in a lithium ion battery, the capacity of a battery cell is determined by the size of the positive electrode and the amount of electrode active material loaded thereon, so the size of the positive electrode is used, but is not limited thereto. For example, when the capacity of a battery cell is determined by the size of the negative electrode and the amount of electrode active material loaded thereon according to a design, the predicted capacity estimation unit 120 can estimate the theoretical capacity of the battery cell based on the size of the negative electrode 220 and the amount of electrode active material corresponding thereto.

[0039] The actual capacity measuring unit 130 measures the actual capacity of the completed battery cell. The capacity of the battery cell may be measured by measuring the current of the cell after a charge / discharge process, but the measurement time and method are not particularly limited.

[0040] The controller 140 can calculate the difference between the predicted capacity of the battery cell estimated by the predicted capacity estimation unit 120 and the actual capacity of the battery cell measured by the actual capacity measurement unit 130, and determine process factors that affect the capacity difference using a trained artificial intelligence model. For this purpose, the controller 140 obtains distribution data of actual capacity against predicted capacity for a plurality of battery cells manufactured through a process.

[0041] FIG. 6 is a graph showing distribution data of measured actual capacity versus predicted capacity for multiple battery cells. Referring to FIG. 6, the horizontal axis of the graph indicates the predicted capacity estimated based on the vision data of the mono cell, and the vertical axis indicates the actual capacity measured by inspection of the completed battery cell, with each point indicating the predicted capacity and actual capacity of the battery cell. As shown in the figure, even for battery cells that have undergone the same manufacturing process, not only do the predicted capacities differ depending on the size of the mono cell electrodes, but the actual capacity also has a large or small error due to various process factors that have an impact as the battery cells undergo multiple processes. In particular, if the actual capacity is significantly lower than the predicted capacity (i.e., the points located relatively low on the graph), this can lead to a deterioration in product quality, so it is necessary to find and manage the process factors that affect such results.

[0042] To determine process factors that affect the difference between the actual capacity and the predicted capacity, the controller 140 can use a clustering algorithm and / or a first artificial intelligence model to classify the battery cells into two or more groups according to a previously set classification criterion (e.g., according to the magnitude of the actual capacity relative to the predicted capacity of the battery cells on the distribution data).

[0043] According to an embodiment, the clustering algorithm may be a density-based spatial clustering of applications with noise (DBSCAN) algorithm. Here, the DBSCAN algorithm is an algorithm for classifying each point indicating the predicted capacity and the actual capacity of the battery cell into a core point, a border point, or a noise point based on whether a specified number of points are included within a specified radius from each of the points. The core point may be a point that includes a specified number of points or more within the radius. The border point may be a point that includes less than a specified number of points within the radius, and the included points include the core point. The noise point may be a point that includes less than a specified number of points within the radius, and the included points do not include the core point. Here, the size and number of the radius may be predefined by a user. According to an embodiment, two or more groups classified according to the size of the actual capacity relative to the predicted capacity of the battery cell on the distribution data may be clusters composed of different core points and border points. Also, the noise point may not be included in two or more groups. 7, the points classified as "GOOD" may be a first cluster consisting of core points and boundary points, and the points classified as "NOT GOOD" may be a second cluster consisting of other core points and other boundary points. Here, the first cluster and the second cluster may be referred to as two or more groups.

[0044] According to an embodiment, the first artificial intelligence model is a machine learning-based classification model and can discontinuize and / or classify adjacent data to improve classification performance. For example, the first artificial intelligence model can utilize a model based on DBSCAN, but is not limited thereto. For example, when the first artificial intelligence model is based on DBSCAN, the first artificial intelligence model can be made to perform supervised learning based on the training data. For example, the first artificial intelligence model can be trained to minimize the difference between the core points, boundary points, or noise points output by the first artificial intelligence model and the core points, boundary points, or noise points labeled in the training data.

[0045] FIG. 7 illustrates the classification of battery cell capacity distribution data into two groups using an artificial intelligence model according to an embodiment. In Figure 7, points classified as "GOOD" refer to battery cells where the difference between the theoretically estimated predicted capacity and the actual capacity is relatively small, and points classified as "NOT GOOD" refer to battery cells where the difference between the predicted capacity and the actual capacity is relatively large. As mentioned above, if the difference between the predicted capacity and the actual capacity is large, it can lead to a decrease in product quality, so it is necessary to find and manage the process factors that affect such results.

[0046] The controller 140 can use the second artificial intelligence model to analyze the distribution data belonging to each group (e.g., a group classified as "GOOD" or a group classified as "NOT GOOD") and the process data of the corresponding battery cell, thereby calculating the importance indicating the effect that each process factor has on the capacity of the battery cell.

[0047] According to one embodiment, the second artificial intelligence model can be trained to calculate the importance of each process factor based on the change in the distribution data due to the change or presence of the process factor for two or more preset process factors. The second artificial intelligence model can be based on a model utilizing an ensemble-based boosting method (e.g., gradient boost for classification, XGBoost (e.g., XGBClassifier, LightGBMClassifier), etc.). Here, the ensemble may be a technique of combining multiple models (e.g., decision trees) to generate a prediction result.

[0048] In one embodiment, the second artificial intelligence model can combine decision trees for each of a plurality of process factors to grasp the importance between the process factors, where each decision tree includes two leaves branching from one decision node.

[0049] In one embodiment, the second artificial intelligence model can be trained to minimize the difference value (i.e., pseudo-residual (or error) value) between the capacity difference of the battery cell classified as the final leaf and the predicted value of the final leaf. Here, the second artificial intelligence model can be configured as a decision tree of a specified depth. The specified depth can be determined as an integer between 8 and 32.

[0050] The second artificial intelligence model can be trained using the process data of the first to third processes as input data and the capacity difference as a predicted value. The second artificial intelligence model can be trained by a process of calculating pseudo residuals for each leaf of the current decision tree, a process of generating a child decision tree of the current decision tree, and a process of generating predicted values ​​for each leaf included in the child decision tree.

[0051] Each leaf can exhibit a pseudo-residual (or error) value, where the pseudo-residual of each leaf of the nth decision tree is the product of the capacity difference between the battery cells classified into each leaf of the nth decision tree and the prediction (P n-1 ) can be calculated based on the difference value.

[0052] Based on the pseudo residual of each leaf of the n-th decision tree, a predicted value for the capacity difference of the battery cells included in each leaf of the n-th decision tree may be calculated. The initial predicted value (P0) may be set to the average value of the capacity difference of all the battery cells. The predicted value (P n ) is the prediction value (P n-1 ) and the output value of the current decision tree's own leaf (O n ) multiplied by a weight (η (e.g., 0.1)) and the sum (P n =P n-1 +η×O n ) where the output value of the current decision tree itself (O n ) can be calculated based on the pseudo residuals of the battery cells included in the leaf. For example, the output value (O n ) may be the average of the pseudo residuals, i.e., the output value (O n ) is the sum of the pseudo residuals (R=r1+r2+…+r M ) divided by the number of battery cells included in the leaf (M) (S=R / M). For example, the output value of the first leaf of the nth decision tree may be different from the output value of the second leaf.

[0053] A decision tree branching from a previous decision tree may be determined based on a gain value. For example, a decision tree for each of two leaves included in the previous decision tree may be determined based on a gain value. That is, the decision tree for each of the two leaves may be different. For example, a decision tree having the highest gain value among the gain values ​​of a candidate decision tree for classifying the presence or absence of process factor 2, a decision tree for classifying the presence or absence of process factor 3, ..., a decision tree for classifying the presence or absence of process factor N may be determined as the decision tree for the first leaf of the previous decision tree as a decision tree candidate for the first leaf of the previous decision tree. Here, the gain value of each candidate decision tree is calculated based on the similarity score of each candidate decision tree, and the similarity score (S) is calculated by summing up the pseudo residuals of the battery cells included in each leaf included in each decision tree (R=r1+r2+...+r M ) (R^2) divided by the number of battery cells (M) (S=(R^2) / M). Here, the similarity score of one decision tree may include the similarity score of the first leaf (S1) and the similarity score of the second leaf (S2). The gain value may be a value obtained by subtracting the similarity score of the parent node from a value obtained by adding up the similarity score of the first leaf (S1) and the similarity score of the second leaf (S2).

[0054] In one embodiment, the decision node may classify the battery cells according to the presence or absence of each process factor or according to a reference value of each process factor, and classify each battery cell into one of two leaves. For example, the first decision node may inquire about the presence or absence of a process factor 1 in the first process. The first decision node may assign a cell ID in which the process factor 1 exists in the first process to a first leaf of the decision node, and assign a cell ID in which the process factor 1 does not exist in the first process to a second leaf of the first decision node. The second decision node may inquire about the presence or absence of a process factor 2 in the first process. The second decision node may assign a cell ID in which the process factor 2 exists in the first process to a first leaf of the second decision node, and assign a cell ID in which the process factor 2 does not exist in the first process to a second leaf of the second decision node. Here, the first decision node and the second decision node may be connected to each other. For example, the judgment process of the second decision node may be performed on the battery cell assigned to the first leaf of the first decision node. In this case, the first decision node may be a parent node or a previous node of the second decision node.

[0055] In one embodiment, the second artificial intelligence model can calculate the importance of each process factor by a permutation importance method based on the process data classified into two or more groups. Here, the importance of the process factor can be based on the gain value of a decision tree in which the process factor is utilized among the decision trees. For example, the sum of the gain values ​​of the decision tree inquiring about the presence or absence of the first process factor can be evaluated as the importance of the first process factor. However, the type of model and the method of utilization are merely examples and are not limited thereto.

[0056] For example, the second artificial intelligence model can compare the capacity difference (difference between predicted capacity and actual capacity) calculated excluding the intervention of process factor 1 in the process data with the capacity difference calculated excluding the intervention of process factor 2, and determine which factor, process factor 1 or process factor 2, has a greater impact on the difference between predicted capacity and actual capacity. In this manner, the importance of each process factor can be calculated by comparing all process factors included in the process data with the capacity difference (distribution data) with and without the intervention of the process factor.

[0057] In this way, the controller 140 can determine the process factors that affect the capacity of the battery cell based on the importance of each process factor calculated using the artificial intelligence model. Once the process factors are determined, the process factors can be controlled through appropriate feedback to improve the capacity quality of the battery cell.

[0058] According to a further embodiment, the controller 140 can classify the distribution data of the actual capacity of the battery cell against the predicted capacity into three or more groups. Unlike Fig. 7 which classifies the distribution data into two groups, Fig. 8 shows the capacity distribution data of the battery cell classified into three groups ("GOOD", "OK", and "NOT GOOD") using an artificial intelligence model according to an embodiment.

[0059] Specifically, when the magnitude of the actual capacity relative to the predicted capacity of the input battery cell is equal to or greater than a first threshold, the controller 140 classifies the distribution data into a first group, “GOOD”, and when the magnitude of the actual capacity relative to the predicted capacity of the input battery cell is equal to or less than a second threshold, the controller 140 classifies the distribution data into a second group, “NOT GOOD”, and otherwise classifies the distribution data into “OK”.

[0060] Here, the battery cells belonging to the first group ("GOOD") can be considered to have relatively good quality since the actual capacity is larger than the theoretical predicted capacity. Therefore, the controller 140 can determine the process factors that have a significant effect on the capacity difference of the battery cells as positive process factors for the distribution data belonging to the first group. Conversely, the battery cells belonging to the second group ("NOT GOOD") can be considered to have relatively low quality since the actual capacity is smaller than the theoretical predicted capacity. Therefore, the controller 140 can determine the process factors that have a significant effect on the capacity difference of the battery cells as negative process factors for the distribution data belonging to the second group. Furthermore, the capacity quality of the battery cells can be improved by increasing the influence of the positive process factors and decreasing the influence of the negative process factors in the subsequent battery manufacturing process.

[0061] 9 is a flow chart showing a method for analyzing process data of a battery cell according to an embodiment. Each step of the method of the embodiment can be performed by any one or a combination of the components of the battery cell process data analysis system 10 described with reference to FIG. 1, i.e., the process data acquisition unit 110, the predicted capacity estimation unit 120, the actual capacity measurement unit 130, and the controller 140, but the system 10 is not essential for performing the method.

[0062] 9, first, a step of acquiring process data of a battery cell (S100) is performed. According to an embodiment, the process data of the battery cell may include data related to a plurality of processes included in a battery manufacturing process, such as an electrode process, an assembly process, and an activation process.

[0063] In addition, in step S100, integrated process data can be generated by linking each process data associated with different processes using a key index, and the integrated process data can be used to determine process factors that affect the difference between the predicted capacity and the actual capacity of the battery cell, among the process factors associated with different processes. By integrating and managing the process data of a plurality of process steps in this manner, the effect of each process factor on the battery capacity can be easily and accurately understood.

[0064] Next, a step (S200) of estimating a predicted capacity of the battery cell based on the process data is performed. The predicted capacity means a theoretical maximum capacity calculated based on the design variables of the battery cell. According to an embodiment, the step (S200) may include a step of calculating the size of the electrodes of the mono cells based on the measured values ​​of the individual mono cells, and a step of estimating a predicted capacity of a battery cell in which a plurality of individual mono cells are stacked based on the size of the electrodes and the amount of electrode active material loaded on the electrodes. Specifically, the size of the electrodes (e.g., the size of the positive electrode or negative electrode) can be calculated from vision data obtained by photographing the upper and lower parts of the surface of a mono cell (a unit cell composed of a positive electrode, a negative electrode, and a separator), and based on the calculated size, the theoretical capacity that the battery cell can have electrically can be estimated. The process of calculating the size of the positive electrode from the vision data of the mono cell is as described above with reference to FIG. 5. Also, as described above, depending on the design, the theoretical capacity of the battery cell can be estimated based on the size of the negative electrode and the amount of electrode active material corresponding thereto.

[0065] Next, the actual capacity of the completed battery cell is measured (S300). As described above, the capacity of the battery cell can be measured by measuring the current of the cell after a charge / discharge process, but the measurement time and method are not particularly limited.

[0066] Next, a step of calculating a difference between the actual capacity and the predicted capacity of the battery cell (S400) is performed. The calculated capacity difference value can be used to determine main process factors that affect the capacity difference of the battery cell in the next step.

[0067] Next, a step (S500) of determining process factors that affect the difference in capacity is performed using the trained artificial intelligence model. Even for battery cells that have undergone the same manufacturing process, not only will the predicted capacity differ depending on the size of the monocell electrode and the amount of electrode active material loaded, but the actual capacity will also have a large or small error due to various process factors that have an effect as the battery cells go through multiple processes. In particular, if the actual capacity is significantly lower than the predicted capacity, this can lead to a deterioration in product quality, so it is necessary to find and manage the process factors that affect such results.

[0068] FIG. 10 is a flow chart showing in more detail the steps of determining the process parameters in FIG. Referring to FIG. 10, the step (S500) may include a step (S510) of acquiring distribution data of actual capacity relative to predicted capacity for a plurality of battery cells manufactured through a process, a step (S520) of classifying the distribution data into two or more groups according to the magnitude of actual capacity relative to predicted capacity of the input battery cells using a first artificial intelligence model, and a step (S530) of calculating the importance indicating the influence of each process factor on the capacity of the battery cells by analyzing the distribution data belonging to each group and the process data of the corresponding battery cells using a second artificial intelligence model.

[0069] According to an embodiment, the first artificial intelligence model is a machine learning-based classification model, and can discontinuize adjacent data to improve classification performance. For example, DBSCAN can be used, but is not limited thereto. In addition, the second artificial intelligence model can be trained to calculate the importance of each process factor based on the degree of change in the distribution data due to the change in the process factor for two or more pre-set process factors. More specifically, the importance of each process factor can be calculated by comparing all process factors included in the process data with the capacity difference (distribution data) due to the presence or absence of the intervention of the process factor. The second artificial intelligence model can find process data classified into two or more groups using a classification model such as XGBClassifier or LightGBMClassifier, and calculate the importance of each process factor using a permutation importance method. However, the types of models and the methods of use are merely examples and are not limited thereto.

[0070] In this way, the process factors that affect the capacity of the battery cell can be determined based on the importance of each process factor calculated using the artificial intelligence model, and further, by controlling the process factors through appropriate feedback, the capacity quality of the battery cell can be improved.

[0071] According to a further embodiment, in the step (S520) of classifying the distribution data into two or more groups, if the magnitude of the actual capacity relative to the predicted capacity of the input battery cell is equal to or greater than a first threshold, the distribution data can be classified into a first group (e.g., a group belonging to "GOOD" in FIG. 8 ), and if the magnitude of the actual capacity relative to the predicted capacity of the input battery cell is equal to or less than a second threshold, the distribution data can be classified into a second group (e.g., a group belonging to "NOT GOOD" in FIG. 8 ).

[0072] In step S500 of determining process factors based on the classification results of step S520, process factors that affect the difference between the predicted capacity and the actual capacity of the battery cell for distribution data belonging to a first group (e.g., the group belonging to "GOOD" in FIG. 8) can be determined as positive process factors, and process factors that affect the difference between the predicted capacity and the actual capacity of the battery cell for distribution data belonging to a second group (e.g., the group belonging to "NOT GOOD" in FIG. 8) can be determined as negative process factors. Furthermore, by increasing the influence of the positive process factors and decreasing the influence of the negative process factors in the subsequent battery manufacturing process, the capacity quality of the battery cell can be improved.

[0073] The battery cell data analysis method according to the above embodiment may be implemented as an application or may be implemented in the form of program instructions executable by various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, and the like, either alone or in combination.

[0074] According to the above embodiment, a system and method for analyzing process data of a battery cell are provided, which can find process factors that affect the capacity of a battery cell by artificial intelligence-based data analysis. In addition, by linking data of different processes using a key index, it is possible to integrate and manage a plurality of process factors that affect the battery capacity in each process step.

[0075] By utilizing the proposed system and method, process factors that have a positive impact on the capacity of the battery cells (i.e., making the actual capacity larger than the predicted capacity) and a negative impact on the capacity of the battery cells (i.e., making the actual capacity smaller than the predicted capacity) can be identified, and the capacity quality of the produced battery cells can be improved by managing the process factors.

[0076] The above description is merely an illustrative example of the technical ideas disclosed in this document, and various modifications and variations may be made by a person having ordinary knowledge in the technical field to which the embodiments disclosed in this document pertain, without departing from the essential characteristics of the embodiments disclosed in this document.

[0077] Therefore, the embodiments disclosed in this document are intended to explain, not to limit, the technical ideas disclosed in this document, and such embodiments do not limit the scope of the technical ideas disclosed in this document. The scope of protection of the technical ideas disclosed in this document must be interpreted according to the claims below, and all technical ideas within the scope equivalent thereto are to be interpreted as being included in the scope of rights of this document.

Claims

1. a process data acquisition unit for acquiring process data of the battery cell; a predicted capacity estimation unit that estimates a predicted capacity of the battery cell based on the process data; an actual capacity measuring unit for measuring an actual capacity of the battery cell; a controller that calculates a difference between a predicted capacity of the battery cell and an actual capacity of the battery cell, and determines process factors that affect the difference using a trained artificial intelligence model; A battery cell process data analysis system comprising:

2. 2. The battery cell process data analysis system of claim 1, wherein the predicted capacity estimation unit calculates a size of an electrode of the mono-cell based on a measured value of each mono-cell, and estimates a predicted capacity of a battery cell including a plurality of stacked mono-cells based on the size of the electrode and an amount of electrode active material loaded on the electrode.

3. The controller: Obtaining distribution data of actual capacity against predicted capacity for a plurality of battery cells manufactured through the process; classifying the distribution data into two or more groups according to the magnitude of the actual capacity of the battery cell relative to the predicted capacity using a first artificial intelligence model; The battery cell process data analysis system of claim 1, characterized in that a second artificial intelligence model is used to analyze distribution data belonging to each group and the corresponding battery cell process data to calculate the importance indicating the influence of the process factors on the capacity of the battery cell.

4. The battery cell process data analysis system of claim 3, wherein the second artificial intelligence model is trained to calculate the importance of each process factor for two or more process factors based on changes in the distribution data due to the presence or absence of the process factors.

5. The controller: classifying the distribution data into a first group when the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or greater than a first threshold, and classifying the distribution data into a second group when the magnitude of the actual capacity of the input battery cell relative to the predicted capacity is equal to or less than a second threshold; 5. The battery cell process data analysis system of claim 4, wherein a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell is determined as a positive process factor for the distribution data belonging to the first group, and a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell is determined as a negative process factor for the distribution data belonging to the second group.

6. The process data of the battery cell includes data related to at least one of an electrode process, an assembly process, and an activation process; the process data acquisition unit generates integrated process data by linking each process data associated with different processes using a key index; 2. The battery cell process data analysis system of claim 1, wherein the controller determines a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell among process factors associated with different processes using the integrated process data.

7. acquiring process data of a battery cell; estimating a predicted capacity of the battery cell based on the process data; measuring an actual capacity of the battery cell; calculating a difference between an actual capacity and a predicted capacity of the battery cell; determining process factors that affect the difference using a trained artificial intelligence model; A method for analyzing process data of a battery cell, comprising:

8. The step of estimating a predicted capacity includes: calculating the electrode sizes of the monocells based on the measurements of the individual monocells; and estimating a predicted capacity of a battery cell including a plurality of individual mono-cells stacked on the basis of a size of the electrode and an amount of electrode active material loaded on the electrode.

9. The step of determining the process parameters comprises: Obtaining distribution data of actual capacity relative to predicted capacity for a plurality of battery cells manufactured through the process; classifying the distribution data into two or more groups according to the magnitude of the actual capacity of the battery cell relative to the predicted capacity using a first artificial intelligence model; and calculating an importance level indicating an influence of the process factors on the capacity of the battery cell by analyzing distribution data belonging to each group and the corresponding process data of the battery cell using a second artificial intelligence model.

10. The battery cell process data analysis method according to claim 9, characterized in that the second artificial intelligence model is trained to calculate the importance of each process factor based on a change in the distribution data due to the presence or absence of two or more pre-set process factors.

11. In the step of classifying the distribution data into two or more groups, if a magnitude of an actual capacity relative to a predicted capacity of the input battery cell is equal to or greater than a first threshold, the distribution data is classified into a first group, and if a magnitude of an actual capacity relative to a predicted capacity of the input battery cell is equal to or less than a second threshold, the distribution data is classified into a second group; 11. The method of claim 10, wherein in the step of determining the process factors, a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell for the distribution data belonging to the first group is determined as a positive process factor, and a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell for the distribution data belonging to the second group is determined as a negative process factor.

12. The process data of the battery cell includes data related to at least one of an electrode process, an assembly process, and an activation process; In the step of acquiring the process data, a key index is used to link the process data associated with different processes to generate integrated process data; 8. The method of claim 7, wherein in the step of determining the process factors, a process factor that affects a difference between a predicted capacity and an actual capacity of the battery cell is determined from among process factors associated with different processes using the integrated process data.

Images