Data processing device and method for battery manufacturing process analysis

Abstract

A data processing device for battery manufacturing process analysis according to one embodiment of the present invention may include at least one processor; and a memory for storing at least one instruction executed through the at least one processor. Here, at least one of the above instructions may include: an instruction to collect process data for each process factor related to multiple batteries; an instruction to construct a machine learning-based performance prediction model for predicting battery performance using the above process data for each process factor; an instruction to generate analytical information showing the influence of one or more of the above process factors on the performance prediction values ​​of the performance prediction model; and an instruction to output the generated analytical information via a predefined GUI (Graphical User Interface).

Description

Technical Field

[0001] This application claims the benefit of the filing date of Korean Patent Application No. 10-2023-0190095, filed with the Korean Intellectual Property Office on December 22, 2023, and all of the contents disclosed in the document of the Korean patent application are incorporated herein.

[0002] The present invention relates to a data processing apparatus and method, and more particularly, to a data processing apparatus and method for analyzing a battery manufacturing process.

Background Art

[0003] A secondary battery is a battery that can be reused through charging even after discharge, and can be used as an energy source for small devices such as mobile phones, tablet PCs, and vacuum cleaners, and can also be used as an energy source for medium and large devices such as automobiles and smart grid ESSs (Energy Storage Systems).

[0004] A battery can be manufactured by sequentially performing an electrode manufacturing process, an assembly process, an activation process, etc. During the progress of the battery manufacturing process and after the completion of the battery manufacturing process, various battery performance indicators such as the resistance, capacity, and full charge time of the battery are considered.

[0005] Various process factors related to each unit process affect the performance of the battery, and since such process factors have a very complex correlation, it is very difficult to grasp which process factors mainly affect the performance of the battery and how much influence the corresponding process factors have.

[0006] As a technology for solving the above problems, there is a need for an appropriate data processing process that can identify process factors that affect the performance of a battery and analyze the influence of the corresponding process factors by using data related to the battery manufacturing process.

Summary of the Invention

Problems to be Solved by the Invention

[0007] The objective of the present invention, which aims to solve the above-mentioned problems, is to provide a data processing device for battery manufacturing process analysis.

[0008] Another object of the present invention, in order to solve the above-mentioned problems, is to provide a data processing method that can be performed by such a data processing device. [Means for solving the problem]

[0009] A data processing device for battery manufacturing process analysis according to one embodiment of the present invention for achieving the above objective may include at least one processor; and a memory for storing at least one instruction executed through the at least one processor.

[0010] Here, at least one of the above instructions may include: an instruction to collect process data for each process factor related to multiple batteries; an instruction to construct a machine learning-based performance prediction model for predicting battery performance using the above process data for each process factor; an instruction to generate analytical information showing the influence of one or more of the above process factors on the performance prediction values ​​of the performance prediction model; and an instruction to output the generated analytical information via a predefined GUI (Graphical User Interface).

[0011] Here, the process data may include data related to one or more unit processes among the electrode coating process, electrode rolling process, assembly process, activation process, and EOL (End Of Line) process.

[0012] The instructions for constructing the above performance prediction model may include instructions to remove process data for one or more process factors based on one of the correlation and importance levels of the multiple process factors.

[0013] The instruction for constructing the above performance prediction model may include an instruction for constructing a performance prediction model that uses the above process factor-specific process data as training data and outputs predicted values ​​for one or more performance factors among battery capacity and resistance.

[0014] The instructions for constructing the above performance prediction model may include instructions for generating multiple performance prediction models by applying different learning algorithms to each other, and instructions for selecting the performance prediction model that shows the best performance from among the performance prediction models based on the difference between the predicted performance value and the measured performance value for each of the above performance prediction models.

[0015] The command that generates the above analysis information may include a command to calculate an influence index for each process factor based on the change in performance prediction values ​​due to changes in individual process factors.

[0016] The command to output the above analysis information via the GUI may include a command to visualize and output one or more of the length, size, direction, and hue of the object corresponding to the process factor, in accordance with the influence index of the relevant process factor.

[0017] The command to output the above analysis information via the GUI may include a command to select the top N process factors (where N is a natural number greater than or equal to 2) with the highest influence index, and a command to visualize and output the objects corresponding to each of the N process factors in accordance with their respective influence indexes.

[0018] The commands that output the above analysis information via the GUI may include commands to group the above-mentioned process factors based on a unit process; commands to sum the influence indices of the process factors included in each unit process; and commands to visualize and output the objects corresponding to each unit process in accordance with the influence indices for each unit process.

[0019] The command that outputs the above analysis information via the above GUI may include a command that outputs the change in performance prediction values ​​due to changes in process factors in the form of a two-dimensional graph.

[0020] A data processing method for battery manufacturing process analysis according to one embodiment of the present invention for achieving the above-mentioned objective may include the steps of: collecting process data for each process factor relating to multiple batteries; constructing a machine learning-based performance prediction model for predicting battery performance using the process data for each process factor; generating analytical information showing the influence of one or more of the process factors on the performance prediction values ​​of the performance prediction model; and outputting the generated analytical information via a predefined GUI (Graphical User Interface).

[0021] Here, the process data may include data related to one or more unit processes among the electrode coating process, electrode rolling process, assembly process, activation process, and EOL (End Of Line) process.

[0022] The step of constructing the above performance prediction model may include the step of removing process data for one or more process factors based on one of the correlation and importance levels of the multiple process factors.

[0023] The step of constructing the above performance prediction model may include the step of constructing a performance prediction model that uses the above process factor-specific process data as training data to output predicted values ​​for one or more performance factors among battery capacity and resistance.

[0024] The steps of constructing the above performance prediction model may include: generating multiple performance prediction models by applying different learning algorithms to each other; and selecting the performance prediction model that exhibits the best performance from among the performance prediction models based on the difference between the predicted performance value and the measured performance value for each of the above performance prediction models.

[0025] The step of generating the above analysis information can include the step of calculating an influence index for each of the process factors based on the change in the performance prediction value due to the change in the individual process factors.

[0026] The step of outputting the above analysis information via the above GUI can include the step of visualizing and outputting one or more of the length, size, direction, and hue of the object corresponding to the process factor so as to correspond to the influence index of the corresponding process factor.

[0027] The step of outputting the above analysis information via the above GUI can include the step of selecting the top N (N is a natural number of 2 or more) process factors with high influence indices; and the step of visualizing and outputting the objects corresponding to each of the N process factors so as to correspond to their respective influence indices.

[0028] The step of outputting the above analysis information via the above GUI can include the step of grouping the plurality of process factors based on the unit process; the step of summing the influence indices of the process factors included in each unit process; and the step of visualizing and outputting the objects corresponding to each unit process so as to correspond to the influence index for each unit process.

[0029] The step of outputting the above analysis information via the above GUI can include the step of outputting the change in the performance prediction value due to the change in the process factor in the form of a two-dimensional graph.

Advantages of the Invention

[0030] According to the embodiments of the present invention as described above, it is possible to provide analysis information regarding process factors that affect the performance of the battery and assist in improving the battery manufacturing process.

Brief Description of the Drawings

[0031] [Figure 1] Shows a general battery manufacturing process. [Figure 2]This is a flowchart illustrating the operation of a data processing method according to an embodiment of the present invention. [Figure 3] This is an example of process data related to an embodiment of the present invention. [Figure 4] This is a flowchart illustrating the operation of a performance prediction model construction method according to an embodiment of the present invention. [Figure 5] This is a reference diagram illustrating a data preprocessing method according to an embodiment of the present invention. [Figure 6] This is an example of a user terminal screen illustrating a GUI according to an embodiment of the present invention. [Figure 7] This is an example of a user terminal screen for explaining the analysis information according to an embodiment of the present invention. [Figure 8] This is an example of a user terminal screen for explaining the analysis information according to an embodiment of the present invention. [Figure 9] This is an example of a user terminal screen for explaining the analysis information according to an embodiment of the present invention. [Figure 10] This is an example of a user terminal screen for explaining the analysis information according to an embodiment of the present invention. [Figure 11] This is a block diagram of a data processing device according to an embodiment of the present invention. [Modes for carrying out the invention]

[0032] The present invention can be modified in various ways and has many embodiments; therefore, specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this should be understood not as limiting the present invention to specific embodiments, but rather as including all modifications, equivalents, and substitutions that fall within the spirit and technical scope of the present invention. Similar reference numerals are used for similar components in the description of each drawing.

[0033] Terms such as First, Second, A, B, etc., may be used to describe various components, but the components should not be limited by such terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the First component may be named the Second component, and similarly, the Second component may be named the First component. The term "and / or" includes a combination of multiple related items or one of multiple related items.

[0034] When it is stated that one component is "linked" or "connected" to another component, it should be understood that this may mean that it is directly linked or connected to that other component, but that there may also be another component in between. Conversely, when it is stated that one component is "directly linked" or "directly connected" to another component, it should be understood that there is no other component in between.

[0035] The terms used in this application are used solely to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless they are clearly different in context. In this application, terms such as “includes” or “having” are intended to specify the presence of features, figures, steps, actions, components, parts, or combinations thereof as described in the specification, and should not be understood to preemptively exclude the presence or possibility of adding one or more other features, figures, steps, actions, components, parts, or combinations thereof.

[0036] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as those generally understood by a person of ordinary skill in the art to which this invention pertains. Terms as defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art, and not as ideal or overly formal unless explicitly defined herein.

[0037] Figure 1 shows a typical battery manufacturing process.

[0038] Batteries can be manufactured by sequentially carrying out multiple unit processes. More specifically, the battery manufacturing process can be classified into N unit processes, and a battery can be manufactured by sequentially carrying out processes 1 through N.

[0039] For example, a battery cell can be manufactured by sequentially carrying out unit processes classified as an electrode coating process (first process), an electrode rolling process (second process), an assembly process (third process), an activation process (fourth process), and an EOL (End Of Line) process (fifth process).

[0040] During the process of individual unit processes, or after battery manufacturing is complete, performance tests may be conducted to verify whether the battery exhibits the intended performance. The performance indicators of the battery being tested may include the battery's discharge capacity, charging resistance, and discharge resistance.

[0041] If performance tests reveal that the battery does not meet its intended performance, it will be necessary to redesign the battery or adjust the manufacturing process variables.

[0042] However, various process factors associated with each unit process affect battery performance, and these process factors have very complex correlations, making it difficult to derive the process factors that primarily affect battery performance. Furthermore, even if process factors that affect battery performance are identified, it is extremely difficult to determine how much those process factors actually influence battery performance.

[0043] The present invention relates to a technology for solving such problems, and concerns a data processing device and method that can use data related to the battery manufacturing process to identify process factors that affect battery performance and analyze the influence of those process factors.

[0044] The operation of the data processing apparatus according to the present invention and various embodiments thereof, as well as the data processing methods performed by the data processing apparatus, will be described in detail below with reference to the attached drawings.

[0045] Figure 2 is an operation flowchart of the data processing method according to an embodiment of the present invention.

[0046] The data processing device can collect process data (S210). Here, the process data may include process data for each of the multiple batteries, broken down by process factor.

[0047] Process data can include data related to each of multiple unit processes. Here, a unit process can include one or more of the following: electrode coating process, electrode rolling process, assembly process, activation process, and EOL process.

[0048] Figure 3 shows an example of process data according to an embodiment of the present invention.

[0049] Referring to Figure 3, the data processing device can collect process data classified into multiple data instances. Here, a data instance can refer to process data categorized by process factor for individual batteries. For example, the data processing device can collect process data categorized by process factor for each of 100,000 battery cells.

[0050] The process factors included in the electrode coating process (first unit process) may include one or more of the following: slurry temperature, slurry flow rate, coating gap, coating rate, and coating thickness.

[0051] The process factors included in the electrode rolling process (second unit process) may include one or more of the following: rolling roll gap, rolling pressure, rolling speed, and rolling thickness.

[0052] The process factors included in the assembly process (third unit process) may include one or more of the following: lamination plate temperature, lamination force, lamination roller temperature, and lamination speed.

[0053] The process factors included in the activation process (the fourth unit process) may include one or more of the following: fixture formation temperature, fixture formation pressure, and inter-process waiting time.

[0054] The process factors included in the EOL process (the fifth unit process) may include one or more of the following: electrolyte volume, performance measurement temperature, and cell thickness.

[0055] The data processing device can use the collected process data to calculate process data related to specific process factors. For example, the data processing device can use the collected process data to calculate process data related to the rolling rate, the thickness of the negative electrode after activation, and the porosity of the negative electrode after activation.

[0056] On the other hand, the process factors for each unit process shown in Figure 3 are illustrative examples for the purpose of clearly explaining the present invention, and the scope of the present invention is not limited to the types of process factors.

[0057] Referring again to Figure 2, the data processing device can construct a performance prediction model using process data (S220). Here, the performance prediction model can be a machine learning-based artificial intelligence model that outputs predicted values ​​for one or more battery performance factors when specific process data is input.

[0058] Specifically, the data processing device can construct a performance prediction model that outputs predicted values ​​for one or more performance factors using process data categorized by process factor. Here, the data processing device can train the performance prediction model using process data categorized by process factor as training data. The performance prediction model can be defined to output predicted values ​​for the battery's performance factors as output data when specific process data is input. Here, the performance prediction values ​​output through the performance prediction model can correspond to predicted values ​​for performance factors that include one or more of the battery's capacity and resistance.

[0059] The data processing device can use the performance prediction model constructed in S220 to generate analytical information showing the influence of one or more process factors on the performance prediction values ​​of the performance prediction model (S230). Here, the analytical information may include influence indices for one or more process factors. An influence index can mean a numerical value that indicates the degree of influence that an individual process factor has on the performance prediction values ​​of the performance prediction model.

[0060] In the embodiment, the data processing device can calculate an influence index on individual process factors based on the change in performance prediction values ​​due to changes in individual process factors.

[0061] For example, a data processing device can calculate an influence index for a given process factor based on the difference between the performance prediction value of a performance prediction model and the average performance prediction value calculated by randomly changing the characteristic value of a specific process factor.

[0062] Another example is that a data processing device can calculate an influence index for a particular process factor based on the difference between the performance prediction value of a performance prediction model and the performance prediction value calculated by removing the characteristic values ​​of that particular process factor.

[0063] In another embodiment, the data processing device can calculate an influence index for a specific process factor based on the contribution of that process factor, which is calculated by propagating the performance prediction values ​​backward to the performance prediction model.

[0064] The influence index can be calculated as a positive or negative number. Here, a positive influence index means that it has a positive impact on the performance forecast (i.e., contributes to an increase in the performance forecast), and a negative number means that it has a negative impact on the performance forecast (i.e., contributes to a decrease in the performance forecast).

[0065] The data processing device can output the analysis information generated in S230 via a predefined GUI (Graphical User Interface) (S240).

[0066] The data processing device can select the top N process factors with the highest influence indices (where N is a natural number greater than or equal to 1, either predefined or entered by the user), and output the selected process factors and their respective influence indices via a GUI.

[0067] The data processing device can generate visual content corresponding to output items entered by the user and output the visual content via a GUI. Here, the visual content can consist of a combination of objects that the user can visually recognize (e.g., lines, arrows, shapes, N-dimensional graphs, etc.). For example, if the user enters specific output items (e.g., impact by instance, impact by process factor, impact by unit process), the data processing device can collect and process analytical information related to the entered output items to generate visual content and output the visual content via a GUI.

[0068] The data processing device can visualize and output one or more of the length, size, direction, and hue of the object corresponding to a process factor, in accordance with the influence index of that process factor. For example, if a request for output of the influence of each process factor is received, the data processing device can visualize and output the influence index for each of the multiple process factors in the form of a horizontal bar graph. In this case, the data processing device can group the process factors by unit process and output each group with a different hue.

[0069] Through the analysis information output via the GUI according to the present invention, users can identify process factors that affect battery performance or intuitively understand the extent to which process factors affect battery performance.

[0070] Figure 4 is an operational flowchart of the performance prediction model construction method according to an embodiment of the present invention. Figure 5 is a reference diagram for explaining the data preprocessing method according to an embodiment of the present invention.

[0071] The data processing device can execute a predefined preprocessing process on the process data collected in S210 (S410).

[0072] The preprocessing process may include a data engineering process that modifies the data instances and a feature engineering process that modifies the process factors contained within each data instance.

[0073] First, in the data engineering process, one or more data instances can be removed according to predefined criteria.

[0074] For example, a data processing device can remove data instances from a pool of 100,000 data instances that pertain to batteries manufactured in a predefined coating lot (LOT) and batteries with feature values ​​that deviate from predefined domain knowledge.

[0075] Next, in the feature engineering process, process data relating to one or more process factors can be removed according to predefined criteria. Specifically, the data processing device can remove process data relating to one or more process factors from among multiple process factors based on one of the correlation and importance levels of those process factors.

[0076] For example, as shown in Figure 5, the data processing device can calculate the correlation coefficients between process factors (#1 to #30) and remove process data for process factors with correlation coefficients exceeding a predefined threshold. Here, the correlation coefficient can include the Pearson correlation coefficient. The data processing device can also calculate an importance index for each process factor and remove process data for process factors with an importance index below a predefined threshold. Here, the importance index can include the permutation importance score.

[0077] Referring again to Figure 4, the data processing device can construct a machine learning-based performance prediction model using the process data after the preprocessing process is complete (S420). Here, the data processing device can generate multiple performance prediction models by applying different learning algorithms to each other.

[0078] For example, the data processing device can generate a Random Forest-based performance prediction model, an XGBoost-based performance prediction model, and a Deep Neural Network-based performance prediction model, and train each of these models using process data categorized by process factor as training data.

[0079] Subsequently, the data processing device calculates a model evaluation index for each of the performance prediction models (S430), and based on the model evaluation index, it can determine which performance prediction model exhibits the optimal performance (S440). Here, the model evaluation index can be calculated based on the difference between the predicted performance value and the measured performance value.

[0080] In this embodiment, the data processing device can calculate RMSE (Root Mean Square Error) and R^2 (R Squared Score) for each of the following performance prediction models: a Random Forest-based performance prediction model, an XGBoost-based performance prediction model, and a Deep Neural Network-based performance prediction model, based on the difference between the predicted performance value and the measured performance value. The data processing device can then determine the performance prediction model with the lowest RMSE or the R^2 value closest to 1 as the optimal performance prediction model.

[0081] For example, if a Random Forest-based charging resistance prediction model has an RMSE of 0.441 and an R² of 0.625, an XGBoost-based charging resistance prediction model has an RMSE of 0.311 and an R² of 0.814, and a Deep Neural Network-based charging resistance prediction model has an RMSE of 0.395 and an R² of 0.669, the data processing device can determine the XGBoost-based charging resistance prediction model with the lowest RMSE and an R² closest to 1 as the optimal charging resistance prediction model.

[0082] In this embodiment, the data processing device can perform an optimization process on the optimal performance prediction model determined in S440. Here, the data processing device can use a Bayesian optimization algorithm to derive the optimal combination of multiple hyperparameters included in the performance prediction model.

[0083] Subsequently, the data processing device generates analytical information to identify process factors affecting battery performance using an optimized performance prediction model (S230), and can output the analytical information via a GUI (S240).

[0084] Figure 6 is an example of a user terminal screen illustrating a GUI according to an embodiment of the present invention.

[0085] Referring to Figure 6, the output items entered by the user and the corresponding analysis information can be visualized and output via a GUI. Here, the GUI may include an output item selection window 610 and an analysis information output window 620.

[0086] The data processing device can receive selection signals for output items entered by the user through the output item selection window 610. Here, the output items may include one or more of the following: influence by instance, influence by process factor, and influence by unit process.

[0087] When a user inputs a specific output item, the data processing device can collect and process analytical information related to the input output item to generate visual content, and output the visual content through the analytical information output window 620.

[0088] If the impact level per instance is selected as an output item and a specific data instance (e.g., a battery identifier) ​​is entered, the data processing device can visualize and output the influence index for each process factor for that data instance (the battery in question).

[0089] If the influence of each process factor is selected as an output item, the data processing device can visualize and output the influence index for each of the multiple process factors.

[0090] If the impact of each process factor is selected as an output item, and a specific process factor (e.g., coating speed) is input, the data processing device can visualize and output the change in the predicted performance value due to the change in the relevant process factor.

[0091] If the impact level for each unit process is selected as an output item, the data processing device can group the process factors by unit process and visualize and output the influence index for each unit process.

[0092] Figures 7 to 10 are examples of user terminal screens used to explain the analysis information according to embodiments of the present invention.

[0093] As shown in Figure 7, if the user selects the impact level per instance as an output item and a specific data instance (e.g., bat #266) is input, the data processing device can visualize and output the influence index for each process factor for that data instance (bat #266).

[0094] Here, the data processing device can select the top N process factors with the highest influence index (where N is a natural number greater than or equal to 2, either predefined or entered by the user), and visualize and output each of the selected process factors and their corresponding objects, corresponding to their respective influence indexes.

[0095] For example, as shown in Figure 7, the data processing device can select the top 7 process factors with the highest influence indices (#10, #15, #7, #4, #19, #27, #21), and output each process factor and its corresponding object (arrow) in descending order. Here, each object (arrow) has a length corresponding to the magnitude of the influence index and can indicate a direction corresponding to the sign of the influence index.

[0096] Furthermore, the data processing device can define the average predicted value of the performance prediction model as the starting point, using the x-axis as a reference, and calculate the start and end points of each object based on the influence index of the process factors, which can then be visualized as shown in Figure 7. This allows the user to intuitively recognize the main process factors that affect the performance of a specific selected battery, and the impact (magnitude and direction) that each of these process factors has on the battery performance.

[0097] As shown in Figure 8, if the user selects the impact of each process factor as an output item, the data processing device can visualize and output the influence index for each of the multiple process factors.

[0098] Here, the data processing device can select the top N process factors with the highest influence index, and visualize and output each of the selected process factors and their corresponding objects in accordance with their respective influence indexes.

[0099] For example, as shown in Figure 8, the data processing device can select the top seven process factors with the highest influence index (#10, #15, #7, #4, #19, #27, #21), and output each process factor and its corresponding object (bar) in descending order. Here, each object (bar) can be represented by a length corresponding to the absolute value of the influence index. This allows the user to intuitively understand the main process factors that affect battery performance and the level of influence each of these process factors has on battery performance.

[0100] Referring to Figure 9, if the user selects the impact of each process factor as an output item and a specific process factor (e.g., #4) is entered, the data processing device can visualize and output the change in the predicted performance value due to the change in that process factor.

[0101] For example, as shown in Figure 9, the data processing device can visualize and output the change in performance prediction values ​​due to changes in the feature values ​​of the selected process factor (#4) in the form of a two-dimensional graph. Here, the data processing device can output both a graph of the relationship between feature values ​​and performance prediction values ​​for each data instance, and a graph of the relationship with the average value of the data instance. This allows the user to intuitively recognize the trend of changes in battery performance due to changes in specific process factors, and the location of feature values ​​where battery performance changes abruptly.

[0102] As shown in Figure 10, if the user selects the impact level for each unit process as an output item, the data processing device can visualize and output the impact index for each unit process.

[0103] Here, the data processing device can group multiple process factors based on a unit process, sum the influence indices of the process factors included in each unit process, and then visualize and output each unit process and its corresponding object in accordance with the influence indices for each unit process.

[0104] For example, as shown in Figure 10, the data processing device can group process factors into electrode coating process, electrode rolling process, assembly process, activation process, and EOL process, sum the influence indices of the process factors included in each unit process, and then visualize and output each unit process and its corresponding object (bar) in accordance with the influence index for each unit process. Here, each object (bar) can be represented by a length corresponding to the absolute value of the influence index for each unit process. This allows the user to intuitively understand the impact of each unit process on battery performance.

[0105] Figure 11 is a block diagram of a data processing device according to an embodiment of the present invention.

[0106] A data processing device 1100 for battery manufacturing process analysis according to an embodiment of the present invention may include at least one processor 1110, a memory 1120 for storing at least one instruction executed through the processor, and a transceiver 1130 connected to a network for communication.

[0107] At least one of the above instructions may include: an instruction to collect process data for each process factor related to multiple batteries; an instruction to construct a machine learning-based performance prediction model for predicting battery performance using the above process data for each process factor; an instruction to generate analytical information showing the influence of one or more of the above process factors on the performance prediction values ​​of the performance prediction model; and an instruction to output the generated analytical information via a predefined GUI (Graphical User Interface).

[0108] Here, the process data may include data related to one or more unit processes among the electrode coating process, electrode rolling process, assembly process, activation process, and EOL (End Of Line) process.

[0109] The instructions for constructing the above performance prediction model may include instructions to remove process data for one or more process factors based on one of the correlation and importance levels of the multiple process factors.

[0110] The instruction for constructing the above performance prediction model may include an instruction for constructing a performance prediction model that uses the above process factor-specific process data as training data and outputs predicted values ​​for one or more performance factors among battery capacity and resistance.

[0111] The instructions for constructing the above performance prediction model may include instructions for generating multiple performance prediction models by applying different learning algorithms to each other, and instructions for selecting the performance prediction model that shows the best performance from among the performance prediction models based on the difference between the predicted performance value and the measured performance value for each of the above performance prediction models.

[0112] The command that generates the above analysis information may include a command to calculate an influence index for each process factor based on the change in performance prediction values ​​due to changes in individual process factors.

[0113] The command to output the above analysis information via the GUI may include a command to visualize and output one or more of the length, size, direction, and hue of the object corresponding to the process factor, in accordance with the influence index of the relevant process factor.

[0114] The command to output the above analysis information via the GUI may include a command to select the top N process factors (where N is a natural number greater than or equal to 2) with the highest influence index, and a command to visualize and output the objects corresponding to each of the N process factors in accordance with their respective influence indexes.

[0115] The commands that output the above analysis information via the GUI may include commands to group the above-mentioned process factors based on a unit process; commands to sum the influence indices of the process factors included in each unit process; and commands to visualize and output the objects corresponding to each unit process in accordance with the influence indices for each unit process.

[0116] The command that outputs the above analysis information via the above GUI may include a command that outputs the change in performance prediction values ​​due to changes in process factors in the form of a two-dimensional graph.

[0117] The data processing device 1100 may further include an input interface device 1140, an output interface device 1150, a storage device 1160, and the like. Each component included in the data processing device 1100 can communicate with one another via a bus 1170.

[0118] Here, processor 1110 can mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which the method according to the embodiment of the present invention is performed. Memory (or storage device) can consist of at least one of a volatile storage medium and a non-volatile storage medium. For example, memory can consist of at least one of a read-only memory (ROM) and a random access memory (RAM).

[0119] The operation of the method according to an embodiment of the present invention can be embodied as a computer-readable program or code on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system. Furthermore, computer-readable recording media can be distributed across networked computer systems, allowing computer-readable programs or code to be stored and executed in a distributed manner.

[0120] Some aspects of the present invention have been described in the context of apparatus, but they can also be described by corresponding methods, where a block or apparatus corresponds to a method step or a feature of a method step. Similarly, aspects described in the context of a method can be described by corresponding blocks or items or features of corresponding apparatus. Some or all of the method steps can be carried out by (or using) hardware devices such as, for example, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, one or more of the most important method steps can be carried out by such devices.

[0121] While preferred embodiments of the present invention have been described above with reference to the present invention, those skilled in the art will understand that the present invention can be modified and altered in various ways without departing from the spirit and scope of the invention as set forth in the following claims. [Explanation of Symbols]

[0122] 610: Output Item Selection Window 620: Analysis Information Output Window 1100: Data Processing Device

Claims

1. A data processing device for battery manufacturing process analysis, At least one processor; and Includes memory for storing at least one instruction executed through the at least one processor, The at least one instruction is, A command to collect process data for multiple batteries, broken down by process factor; Instructions to construct a machine learning-based performance prediction model for predicting battery performance using the process data for each process factor; A command to generate analytical information showing the influence of one or more of the aforementioned process factors on the performance prediction values ​​of the performance prediction model; and A data processing device that includes an instruction to output the generated analysis information.

2. The aforementioned process data is The data processing apparatus according to claim 1, which includes data related to one or more unit processes among the electrode coating process, electrode rolling process, assembly process, activation process, and EOL (End Of Line) process.

3. The instruction for constructing the aforementioned performance prediction model is: The data processing apparatus according to claim 2, comprising an instruction to remove process data relating to one or more process factors based on one of the correlation and importance of the process factors relating to the process factors.

4. The instruction for constructing the aforementioned performance prediction model is: The data processing device according to claim 2, which includes an instruction to construct a performance prediction model that uses the process data for each process factor as training data and outputs predicted values ​​for one or more performance factors among battery capacity and resistance.

5. The instruction for constructing the aforementioned performance prediction model is: Instructions for generating multiple performance prediction models by applying different learning algorithms to each other; and The data processing device according to claim 4, comprising an instruction to select the performance prediction model that exhibits the optimal performance from among the performance prediction models based on the difference between the predicted performance value and the measured performance value for each of the performance prediction models.

6. The command to generate the aforementioned analysis information is: The data processing apparatus according to claim 1, comprising an instruction to calculate an influence index for each process factor based on the change in performance prediction values ​​due to changes in individual process factors.

7. The command to output the aforementioned analysis information is: The data processing apparatus according to claim 6, comprising an instruction for visualizing and outputting one or more of the length, size, direction, and hue of an object corresponding to a process factor, in a manner corresponding to the influence index of the process factor.

8. The command to output the aforementioned analysis information is: An instruction to select the top N process factors (where N is a natural number greater than or equal to 2) with the highest influence index; and The data processing apparatus according to claim 6, comprising an instruction for visualizing and outputting objects corresponding to each of the N process factors in accordance with their respective influence indices.

9. The command to output the aforementioned analysis information is: An instruction to group the multiple process factors based on a unit process; An instruction to sum the influence indices of process factors included in each unit process; and The data processing apparatus according to claim 6, comprising an instruction for visualizing and outputting each unit process and its corresponding object in accordance with the influence index for each unit process.

10. The command to output the aforementioned analysis information is: The data processing device according to claim 6, which includes an instruction to output changes in performance prediction values ​​due to changes in process factors in the form of a two-dimensional graph.

11. A data processing method for analyzing the battery manufacturing process, A step of collecting process data for each process factor related to multiple batteries; A step of constructing a machine learning-based performance prediction model for predicting battery performance using the process data for each process factor; A step of generating analytical information showing the influence of one or more of the process factors on the performance prediction value of the performance prediction model; and A data processing method including the step of outputting the generated analysis information.

12. The aforementioned process data is The data processing method according to claim 11, comprising data related to one or more unit processes among the electrode coating process, electrode rolling process, assembly process, activation process, and EOL (End Of Line) process.

13. The step of constructing the aforementioned performance prediction model is: The data processing method according to claim 12, further comprising the step of removing process data relating to one or more process factors based on one of the correlation and importance of the process factors.

14. The step of constructing the aforementioned performance prediction model is: The data processing method according to claim 12, further comprising the step of constructing a performance prediction model that uses the aforementioned process data for each process factor as training data to output predicted values ​​for one or more performance factors among battery capacity and resistance.

15. The step of constructing the aforementioned performance prediction model is: The steps of generating multiple performance prediction models by applying different learning algorithms to each other; and The data processing method according to claim 14, further comprising the step of selecting the performance prediction model that exhibits optimal performance from among the performance prediction models based on the difference between the predicted performance value and the measured performance value for each of the performance prediction models.

16. The step of generating the aforementioned analysis information is: The data processing method according to claim 11, further comprising the step of calculating an influence index for each process factor based on the change in performance prediction values ​​due to changes in individual process factors.

17. The step of outputting the aforementioned analysis information is: The data processing method according to claim 16, comprising the step of visualizing and outputting one or more of the length, size, direction, and hue of an object corresponding to a process factor, so as to correspond to the influence index of the process factor.

18. The step of outputting the aforementioned analysis information is: A step of selecting the top N process factors (where N is a natural number greater than or equal to 2) with the highest influence index; and The data processing method according to claim 16, further comprising the step of visualizing and outputting objects corresponding to each of the N process factors in accordance with their respective influence indices.

19. The step of outputting the aforementioned analysis information is: A step of grouping the multiple process factors based on a unit process; A step of summing the influence indices of process factors included in each unit process; and The data processing method according to claim 16, further comprising the step of visualizing and outputting each unit process and its corresponding object in accordance with the influence index for each unit process.

20. The step of outputting the aforementioned analysis information is: The data processing method according to claim 16, further comprising the step of outputting the change in performance prediction values ​​due to changes in process factors in the form of a two-dimensional graph.

Images