Electronic device and method for controlling same

Abstract

According to an embodiment disclosed in the present document, an electronic device comprises: a communication circuit communicating with a drying device including a plurality of drying zones for drying an input electrode and a plurality of heating devices disposed to respectively correspond to the plurality of drying zones to provide heat to the electrode; a memory for storing one or more instructions; and a processor. The one or more instructions, when executed, may be configured to cause the processor to: acquire driving state information about the drying device when the drying device operates after being floated; acquire state information about at least one electrode put into the drying device; determine an output of each of the plurality of heating devices on the basis of the driving state information about the drying device and the state information about the at least one electrode; and control the plurality of heating devices on the basis of the determined output of each of the plurality of heating devices.

Description

Electronic device and method of controlling it

[0001] This application claims the benefit of priority to Republic of Korea Patent Application No. 10-2024-0104157, dated August 5, 2024, and Republic of Korea Patent Application No. 10-2024-0127271, dated September 20, 2024, the entire contents of which are incorporated herein by reference.

[0002] The embodiments disclosed in this document relate to an electronic device and a method for controlling the same.

[0003] Recently, research and development on secondary batteries has been actively conducted. Here, secondary batteries are rechargeable and include both conventional Ni / Cd and Ni / MH batteries, as well as recent lithium-ion batteries. Among secondary batteries, lithium-ion batteries have the advantage of having a much higher energy density than conventional Ni / Cd and Ni / MH batteries. Furthermore, lithium-ion batteries can be manufactured in a compact and lightweight form, making them suitable for use as power sources for mobile devices. Recently, their use has expanded to include power sources for electric vehicles, attracting attention as a next-generation energy storage medium.

[0004] Secondary batteries include anodes and cathodes, and can be produced through a manufacturing process for the electrodes corresponding to the cathode or anode. The manufacturing process for secondary batteries can involve various processes, including a drying process to remove moisture from the electrodes during the manufacturing process. This drying process requires proper drying of the electrode surface. Failure to meet these drying conditions can result in electrode defects.

[0005] One purpose of the embodiments disclosed in this document is to provide an electronic device and a control method thereof for stably managing an electrode surface in a drying process by optimally controlling the state of a drying device.

[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 descriptions below.

[0007] According to an embodiment disclosed in the present document, an electronic device includes a communication circuit for communicating with a drying device including a plurality of drying zones for drying inserted electrodes and a plurality of heating devices arranged to correspond to each of the plurality of drying zones and providing heat to the electrodes; a memory storing one or more instructions; and a processor, wherein the one or more instructions, when executed, cause the processor to: when the drying device is operated after floating, obtain operating status information of the drying device; obtain status information of at least one electrode inserted into the drying device; determine an output of each of the plurality of heating devices based on the operating status information of the drying device and the status information of the at least one electrode; and control the plurality of heating devices based on the determined output of each of the plurality of heating devices.

[0008] According to an embodiment, the driving status information of the drying device may include the idle time of the drying device, the previous operating time, and the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged, and the status information of the electrode may include the initial temperature of the electrode surface located in each of the plurality of drying zones and the target temperature of the electrode surface at the target time.

[0009] According to an embodiment, the target time point may include a time point at which time elapses from the time point at which the electrode is inserted into the drying device until the inserted electrode reaches a drying zone from which the electrode is discharged among the plurality of drying zones.

[0010] According to an embodiment, the driving status information of the drying device and the status information of the at least one electrode may be input into a learned deep learning model, and the output of each of the plurality of heating devices output from the deep learning model may be checked.

[0011] According to an embodiment, the deep learning model may be a learning model that models a correlation between the learning input data set and the learning output data set by using a plurality of driving status information of the drying device corresponding to a plurality of drivings of the drying device and a plurality of status information of a plurality of electrodes as a learning input data set, and using outputs of each of the plurality of heating devices corresponding to a plurality of drivings of the drying device as a learning output data set.

[0012] According to an embodiment, the learned deep learning model may include an LSTM model.

[0013] According to an embodiment, the processor may select a plurality of variables related to at least one of an operating state of the drying device and a state of the electrode, and analyze a correlation between the plurality of variables and a target temperature of the electrode surface according to a moving speed of the electrode to determine an input variable of the deep learning model.

[0014] According to an embodiment, the processor may control the output of all heating devices when the electrode is a cathode, and may control the output of a heating device corresponding to a drying zone from which the electrode is discharged among the plurality of heating devices when the electrode is an anode.

[0015] According to an embodiment, the plurality of drying zones may include a first drying zone, a second drying zone, and a third drying zone arranged along a movement path of the electrode.

[0016] According to an embodiment disclosed in the present document, a control method of an electronic device may include: when the drying device is operated after floating, obtaining operating status information of the drying device; obtaining status information of at least one electrode inserted into the drying device; determining an output of each of the plurality of heating devices based on the operating status information of the drying device and the status information of the at least one electrode; and controlling the plurality of heating devices based on the determined output of each of the plurality of heating devices.

[0017] The electronic device and its control method according to the embodiments disclosed in this document can stably manage the surface of an electrode during a drying process. Accordingly, electrode defects can be prevented.

[0018] In addition, the electronic device and its control method according to the embodiments disclosed in this document can effectively determine optimal conditions using a learned deep learning model even when the drying device has various states, such as when it is operated after floating.

[0019] In addition, various effects may be provided, either directly or indirectly, through this document.

[0020] FIG. 1 is a block diagram showing the configuration of a system according to one embodiment disclosed in this document.

[0021] FIG. 2 is a drawing showing the configuration of a drying device (200) according to one embodiment disclosed in this document.

[0022] FIG. 3 is a diagram showing an example of the structure of a deep learning model according to one embodiment disclosed in this document.

[0023] FIG. 4 is a diagram showing an example of a result of applying a deep learning model according to one embodiment disclosed in this document.

[0024] FIGS. 5 to 8 are diagrams showing examples of correlation analysis for determining input variables of a deep learning model according to one embodiment disclosed in this document.

[0025] Figure 9 is a flowchart for explaining a control method according to one embodiment disclosed in this document.

[0026] FIG. 10 is a flowchart for explaining a specific control method of an electronic device according to one embodiment disclosed in this document.

[0027] FIG. 11 is a flowchart illustrating a method for obtaining a learned deep learning model according to an embodiment disclosed in this document.

[0028] Hereinafter, various embodiments of the present invention will be described with reference to the attached drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that the present invention encompasses various modifications, equivalents, and / or alternatives of the embodiments.

[0029] In this document, the singular form of a noun corresponding to an item may include one or more of said items, unless the context clearly indicates otherwise. In this document, phrases such as "A or B," "at least one of A and B," "at least one of A or B," "A, B, or C," "at least one of A, B, and C," and "at least one of A, B, or C" may each include any one of the items listed together in that phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish the corresponding element from other corresponding elements, and do not limit the corresponding elements in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as being “coupled” or “connected” to another component (e.g., a second component), with or without the terms “functionally” or “communicatively,” it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0030] Each component (e.g., a module or a program) described in this document may include one or more entities. According to various embodiments, one or more components or operations of the components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0031] The term "module" or "part" used in this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0032] Various embodiments of the present document may be implemented as software (e.g., a program or an application) including one or more instructions stored in a machine-readable storage medium (e.g., memory). For example, a processor of the device may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the device to operate to perform at least one function according to the at least one instruction called. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, "non-transitory" only means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0033]

[0034] FIG. 1 is a block diagram showing the configuration of a system (10) according to one embodiment disclosed in this document. Referring to FIG. 1, the system (10) may include a drying device (200) for drying an electrode and an electronic device (100) for communicating with and controlling the drying device (200).

[0035] The electrode manufacturing process may include various types of processes, and for example, a drying process for drying the surface of the electrode. For example, the electrode drying process may be performed after a slitting process and a notching process. Here, the slitting process may refer to a process for cutting the width of the electrode to meet the battery specifications, and the notching process may refer to a process for removing the remaining portion of the non-coated region where the active material is not applied, except for the portion for grounding the electrode tab. In the drying process, heat may be applied to the surface of the electrode after the notching process to remove moisture from the electrode.

[0036] Insufficient or excessive drying of the electrode during the drying process can affect its quality and lead to defects. For example, if the electrode is not sufficiently dried, moisture contained within the electrode can cause defects. Furthermore, if the electrode is overdried or dried at excessively high temperatures, surface defects such as heat wrinkles can occur.

[0037] In particular, in the case where the drying device (200) for drying the electrode in the drying process is operated after floating, the state of the drying device (200) and the electrode may vary each time the drying device (200) is operated after floating, and it is difficult to perform appropriate drying for each state.

[0038] Accordingly, the electronic device (100) can determine and control the state of the drying device (200) during the electrode drying process. Through this, the electrode surface can be stably managed and defects can be prevented during the electrode drying process. In particular, the electronic device (100) can control the drying device (200) to an optimal state by considering various initial states, even when the drying device (200) is operated after floating.

[0039] The drying device (200) may include a plurality of drying zones (210) and a plurality of heating devices (220). In one embodiment, the plurality of drying zones may include a first drying zone, a second drying zone, and a third drying zone, which are arranged along a movement path of the electrodes. For example, the first drying zone may be an input zone where the electrodes are introduced into the drying device (200), the third drying zone may be an output zone where the electrodes are discharged outside the drying device (200), and the second drying zone may be an intermediate zone.

[0040] The drying device (200) can have electrodes inserted therein. The drying device (200) can provide heat to the electrodes inserted therein, thereby drying the surface of the electrodes. The electrodes can be transported in a predetermined direction along a transport path, and the surface can be dried while being transported in the predetermined direction within the drying device (200).

[0041] To this end, the drying device (200) may include a plurality of heating devices (220). The plurality of heating devices (220) may be arranged to correspond to each of the plurality of drying zones and may provide heat to the electrodes. For example, the heating device may be an MIR lamp that emits infrared rays toward the electrodes. However, this is merely an example and the type of heating device is not limited thereto.

[0042] The electronic device (100) can manage the electrodes during the drying process, determine the state of the drying device (200) to dry under optimal conditions, and control the drying device (200). For example, when the electronic device (100) is operated after the drying device (200) is floated, the electronic device (100) can determine the output of each of the plurality of heating devices (220) and control the plurality of heating devices (220) based on the determined output.

[0043] According to one embodiment, the electronic device (100) may include a communication circuit (110), a memory (120), and a processor (130). At least one of the components included in the electronic device (10) may be omitted, or another component may be added to the electronic device (10). Additionally or alternatively, some of the components may be implemented in an integrated manner, or may be implemented as a single or multiple entities. At least some of the components within the electronic device (10) may be implemented in an integrated manner, or may be implemented as a single or multiple entities. At least some of the components within the electronic device (10) may be connected to each other via a bus, a general purpose input / output (GPIO), a serial peripheral interface (SPI), or a mobile industry processor interface (MIPI), and may exchange data and / or signals.

[0044] The communication circuit (110) can establish a wired or wireless communication channel with an external device (e.g., the drying device (100)) and transmit and receive various data with the external device. The communication circuit (110) can include at least one port for connecting to the external device via a wired cable in order to communicate with the external device via a wire. The communication circuit (110) can be configured to be connected to a cellular network (e.g., 3G, LTE, 5G, Wibro, or Wimax) by including a cellular communication module. According to one embodiment, the communication circuit (110) can transmit and receive data with the external device using short-range communication (e.g., Wi-Fi, Bluetooth, Bluetooth Low Energy (BLE), UWB) by including a short-range communication module, but is not limited thereto. For example, the communication circuit (110) of the electronic device (100) can communicate with the drying device (200) and exchange various data and / or signals through the communication.

[0045] The memory (120) can store various data used by at least one component (e.g., the processor (130)). The memory (120) can store instructions for the operation of the processor (130). A program can be stored as software in the memory (120) and may include, for example, an operating system, middleware, or an application. Unless otherwise specified, the memory (120) in the present disclosure may refer to a set of one or more memories (120). For example, the memory (120) can store a trained deep learning model.

[0046] The processor (130) is a component that can perform calculations or data processing related to control and / or communication of each component of the electronic device (10), and can be operatively connected to the components of the electronic device (10). The processor (130) can load commands or data received from other components of the electronic device (10) into the memory (120), process the commands or data stored in the memory (120), and store the resulting data. Unless there are special circumstances, the processor (130) in the present disclosure may mean a set of one or more processors (130).

[0047] According to one embodiment, the processor (130) can obtain operating status information of the drying device (200) when the drying device (200) is operated after floating. When the drying device (200) is operated after floating, the drying device (200) can have various operating states, and the processor (130) can obtain operating status information of the drying device (200) in order to optimally manage drying of the electrode according to the operating state of the drying device (200). For example, the processor (130) can obtain operating status information of the drying device (200) by communicating with the drying device (200) through the communication circuit (110).

[0048] According to one embodiment, the operating state information of the drying device (200) may include at least some of the floating time of the drying device (200), the previous operating time, the initial state of each of the plurality of drying zones (210), and the initial state of each of the plurality of heating devices (220). Here, the floating time of the drying device (200) may refer to the time between the end time of the previous operation and the start time of the current operation, and the previous operating time of the drying device (200) may refer to the duration of the previous operation. The initial state of each of the plurality of drying zones (210) may include, for example, the internal temperature of each drying zone. The initial state of each of the plurality of heating devices (220) may include, for example, the initial temperature of each of the heating devices. In addition, the initial state here may refer to the state at the time when the drying device (200) is operated after floating.

[0049] More specifically, the operating status information of the drying device (200) may include the idle time of the drying device, the previous operating time, and the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged.

[0050] According to one embodiment, the processor (130) can obtain status information of at least one electrode inserted into the drying device (200). Since the status of the electrode may vary when the drying device (200) is operated after floating, the processor (130) can obtain status information of the electrode.

[0051] According to one embodiment, the status information of the electrode may include an initial temperature of the electrode surface located in each of the plurality of drying zones (220) and a target temperature of the electrode surface at a target time.

[0052] The processor (130) aims to stably manage the surface of the electrode during the electrode drying process, and for this purpose, may aim to control the surface temperature of the electrode within a target temperature range. Accordingly, when the drying device (200) is operated after floating to control the electrode drying process, the processor (130) may obtain the initial temperature of the electrode surface and the target temperature of the electrode surface at the target time as electrode status information.

[0053] In one embodiment, the target time point may include a time point at which time it takes for the electrode to reach a drying zone from among the plurality of drying zones (210) from which the electrode is discharged, from the time point at which the electrode is inserted into the drying device (200). That is, the target time point may include a time point at which time it takes for the electrode to reach the last drying zone, from the time point at which the electrode is inserted into the drying device (200). For example, if the plurality of drying zones (220) include a first drying zone, a second drying zone, and a third drying zone, the target time point may be a time point at which time it takes for the electrode to reach the third drying zone, from the time point at which the electrode is inserted into the drying device (200). For example, if the electrode is inserted into the first drying zone and reaches the third drying zone 9 seconds later, the target time point may be a time point at which 9 seconds have passed since the electrode was inserted into the first drying zone.

[0054] Since the processor (130) aims to ensure that the surface temperature of the electrode satisfies the target temperature when the drying process of the electrode is performed, the time elapsed from the time the electrode is inserted into the drying device (200) until the inserted electrode reaches the drying zone from which the electrode is discharged among the plurality of drying zones (210) can be set as the target time.

[0055] In addition, target points in time may be set for each of the plurality of drying zones (220). That is, different target points in time may be set for each drying zone. For example, if the plurality of drying zones (220) include a first drying zone, a second drying zone, and a third drying zone, and the time it takes for an electrode introduced into the drying device (200) to reach the first drying zone is 7 seconds, the time it takes for the electrode to reach the second drying zone is 8 seconds, and the time it takes for the electrode to reach the third drying zone is 9 seconds, the target point in time for the first drying zone may be set to a time when 7 seconds have elapsed since the electrode was introduced, the target point in time for the second drying zone may be set to a time when 8 seconds have elapsed since the electrode was introduced, and the target point in time for the third drying zone may be set to a time when 9 seconds have elapsed since the electrode was introduced. In this way, the processor (130) may set a target point in each drying zone to more precisely manage the drying conditions during the process in which the electrode passes through each drying zone.

[0056] In one embodiment, when the time elapsed from the time the electrode is inserted into the drying device (200) until it reaches the final drying zone is defined as the first target time point, the target time point may further include a second target time point, which is a second time elapsed from the first target time point. The second time point may be preset. For example, the second time point may be set based on the movement speed of the electrode, similar to the first time point.

[0057] The output of the heating device in the final drying zone may be set higher than the output of the heating device in other drying zones, and the electrode surface may be overdried due to the heat provided while the electrode passes through the final drying zone or the residual heat after passing through. Therefore, the processor (130) may set a point in time after a certain period of time has elapsed since the electrode reaches the final drying zone as an additional second target point in time, and manage the target temperature of the electrode surface at the second target point in time. For example, the second target point in time may be a point in time 6 seconds after the first target point in time.

[0058] For example, if the drying device (200) is operated with a low initial temperature of the electrode, and the target temperature at the second target time point is not taken into consideration, the output of the plurality of heating devices (220) may be increased to reach the target temperature of the electrode surface at the first target time point. In this case, the output of the heating devices may be increased in the last drying zone, and the temperature rise slope of the electrode surface may become large. Accordingly, the surface temperature of the electrode may exceed the target temperature range after the first target time point has elapsed.

[0059] Accordingly, according to one embodiment, the processor (130) may further consider the target temperature of the electrode surface at a second target time point, a predetermined amount of time having elapsed from the first target time point. In this case, the target temperature of the electrode surface at the second target time point may be utilized as an input variable (target temperature of the electrode surface at the target time point) of the deep learning model described below.

[0060] According to one embodiment, the processor (130) may determine the output of each of the plurality of heating devices (220) based on the operating status information of the drying device and the status information of at least one electrode. The processor (130) may determine the output of each of the plurality of heating devices (220) for drying management of the electrodes when the drying device (200) is operated after floating.

[0061] In one embodiment, the processor (130) may determine the output of each of the plurality of heating devices (220) using the trained deep learning model. To this end, the processor (130) may generate a deep learning model and train the deep learning model to obtain the trained deep learning model. In some cases, the generation and training of the deep learning model may be performed in a separate device (e.g., an external server for training) other than the electronic device (100). In this case, the processor (130) may obtain the trained deep learning model from the external server.

[0062] Training of deep learning models

[0063] In the following, it is assumed that a deep learning model is created and learned by an electronic device (100).

[0064] First, the processor (130) can generate a deep learning model. The generation of a deep learning model may mean that the structure of the neural network constituting the deep learning model and the initial values ​​of the parameters assigned to the neural network are set.

[0065] In addition, the processor (130) can ultimately generate a learned deep learning model by training the generated deep learning model. Here, generating a learned deep learning model through training means that the initially generated deep learning model is trained using a plurality of training data by a training method, thereby creating a trained deep learning model set to perform a desired characteristic (or purpose). As described above, such training may be performed in the electronic device (100) itself, or may be performed through a separate server and / or system. Examples of training methods for such deep learning models include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0066] The structure of a neural network constituting a deep learning model may be comprised of multiple neural network layers. Each of the multiple neural network layers has multiple weight values ​​and can perform neural network operations through operations between the operation results of the previous layer and the multiple weights. The multiple weights of the multiple neural network layers can be optimized based on the training results of the deep learning model. For example, a loss function or cost function can be set for optimal training of a deep learning model, and the multiple weights can be updated during each training process so that the loss value or cost value obtained from the deep learning model is reduced or minimized.

[0067] The artificial neural networks that make up the deep learning model may include, but are not limited to, a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM) neural network, a restricted boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), or deep Q-networks.

[0068] The processor (130) can acquire a training data set for training a deep learning model. The training data set may be composed of, for example, previously acquired data or data generated based on the acquired data, and an appropriate data set can be configured depending on the training method, purpose, type of neural network, etc. of the deep learning model.

[0069] In one embodiment, a deep learning model can learn the correlation between input data and output data using a training input data set and a training output data set. The deep learning model can be trained to predict output when data is input, and can be trained to predict training output data from training input data. Through this, the deep learning model can model the correlation between training input data and output data during the training process. This can be understood as performing supervised learning of the deep learning model, and the deep learning model can be trained using the training output data set as a label. For example, the deep learning model can be trained by generating predictions from the training input data set and comparing the predictions with the output data set.

[0070] According to one embodiment, the deep learning model may use a plurality of driving status information of the drying device (200) corresponding to a plurality of driving operations of the drying device (200) and a plurality of status information of a plurality of electrodes as a learning input data set, and may use the outputs of each of a plurality of heating devices (220) corresponding to a plurality of driving operations of the drying device (200) as a learning output data set.

[0071] For example, the processor (130) may obtain operating status information of the drying device (200) obtained from each of the previous multiple operations of the drying device (200) and configure it as a learning input data set. In addition, the processor (130) may obtain outputs of each of the plurality of heating devices (220) obtained from each of the previous multiple operations of the drying device (200) and configure it as a learning output data set.

[0072] Since the processor (130) seeks to derive the optimal output of the plurality of heating devices (220) for drying the electrode using a deep learning model, the output of each of the plurality of heating devices corresponding to the plurality of operations of the drying device (200) can be configured as a training output data set of the deep learning model.

[0073] In addition, the processor (130) may select variables related to the output of a plurality of heating devices (220) or the target temperature of the electrode surface to improve the performance of the deep learning model, and configure a learning input data set with data corresponding to the selected variables.

[0074] In the case of a deep learning model, if the relationship between input and output is low or non-existent, the learning performance and the prediction performance of the result will deteriorate, so the processor (130) can select variables related to the target output characteristics to configure a learning input data set.

[0075] According to one embodiment, the processor (130) may select a plurality of variables related to at least one of the operating state of the drying device (200) and the state of the electrode. The processor (130) may select a plurality of variables and analyze them to select input variables for training a deep learning model. For example, the processor (130) may select variables such as the initial state of the drying device (200) and the initial surface temperature of the electrode as the plurality of variables.

[0076] According to one embodiment, the processor (130) can analyze the correlation between a plurality of variables according to the movement speed of the electrode and the target temperature of the electrode surface to determine input variables for learning the deep learning model.

[0077] The electrode can be moved in a predetermined direction within the drying device (200), and the moving speed of the electrode can be set differently depending on the case. In order to optimally control the drying process of the electrode even in environments where the moving speed of the electrode is different, the electronic device (100) can determine variables that have a correlation even at different moving speeds as input variables of the deep learning model.

[0078] To this end, the processor (130) can analyze the correlation and reciprocal relationship between input variables and output characteristics according to the movement speed of the electrode. For example, in the drying process of the electrode, the movement speed of the electrode can be set to 1000 mm / s or 300 mm / s, and the processor (130) can select input variables that have a correlation and reciprocal relationship with the target surface temperature of the electrode as the output characteristic in both the cases where the movement speed of the electrode is 1000 mm / s and 300 mm / s.

[0079] However, this is merely an example, and the training of a deep learning model and the determination of input variables for training may be performed by an external device (e.g., an external server for training) rather than the processor (130). In this case, the processor (130) may obtain the trained deep learning model and input variable information of the deep learning model from the external server via the communication circuit (110).

[0080] According to one embodiment, input variables constituting training input data of a deep learning model may include 1) the idle time of the drying device, 2) the previous operating time of the drying device, 3) the initial temperature of the electrode surface located in each of the plurality of drying zones, 4) the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged, and 5) the target temperature of the electrode surface at the target time.

[0081] The above input variables can be determined through the correlation analysis and reciprocal relationship analysis described above, and the basis for calculating the above input variables will be supplemented and explained in FIGS. 5 to 8.

[0082] For example, the purpose of learning the output of each of the plurality of heating devices using a deep learning model is to manage the drying state of the electrode surface during the drying process of the electrode through the drying device (200), and to control the temperature of the electrode surface to a target temperature for this purpose. Therefore, the processor (130) may include the target temperature of the electrode surface at the target time as an input variable as input data for learning the deep learning model.

[0083] In this case, the deep learning model can be trained by receiving input data for training consisting of input variables of 1) the dead time of the drying device, 2) the previous operating time of the drying device, 3) the initial temperature of the electrode surface located in each of the multiple drying zones, 4) the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged, and 5) the target temperature of the electrode surface at the target time.

[0084] In this way, the processor (130) configures a learning input data set and a learning output data set, and repeatedly trains a deep learning model using the same, thereby obtaining a trained deep learning model optimized for predicting the output of each of the plurality of heating devices (220).

[0085] In one embodiment, the trained deep learning model may include a Long-Short Term Memory (LSTM) model. The LSTM model is a model that demonstrates excellent performance in analyzing time-series data or signal data, and is also a model that excels in learning to identify complex patterns. Therefore, the electronic device (100) may utilize the LSTM model to learn the complex relationships between various variables related to the state of the drying device (200) and the electrodes and the output values ​​of each of the plurality of heating devices (220).

[0086]

[0087] Utilizing trained deep learning models

[0088] According to one embodiment, the processor (130) may input the operating status information of the drying device (200) and the status information of at least one electrode into the learned deep learning model. In addition, the processor (130) may check the output of each of the plurality of heating devices (220) output from the learned deep learning model. That is, the processor (130) may determine the optimal output of the plurality of heating devices (220) when the drying device (200) is operated after being floated, using the learned deep learning model.

[0089] As described above, the processor (130) can verify the output of each of the plurality of heating devices (220) by using the output from the learned deep learning model. The processor (130) can derive the output of each of the plurality of heating devices (220) by inputting information about input variables constituting the learning input data into the learned deep learning model.

[0090] According to one embodiment, the processor (130) inputs 1) the floating time of the drying device (200) obtained at the time of operation after floating, 2) the previous operating time of the drying device, 3) the initial temperature of the electrode surface located in each of the plurality of drying zones, 4) the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged, and 5) the target temperature of the electrode surface at the target time into the learned deep learning model, and can confirm the output of each of the plurality of heating devices (220) output from the learned deep learning model.

[0091] According to one embodiment, the processor (130) can control the plurality of heating devices based on the output of each of the determined plurality of heating devices. This can be understood as the processor (130) controlling the plurality of heating devices (220) in the initial operation state when the drying device (200) is operated after floating.

[0092] The processor (130) can control the output of each of the plurality of heating devices (220) when the drying device (200) is operated after floating. That is, the processor (130) can set the output values ​​derived from the learned deep learning model as the initial output of each of the plurality of heating devices (220).

[0093] Additionally, the processor (130) may monitor the surface temperature of the electrode during the operation of the drying device (200). According to one embodiment, the processor (130) may correct the output of at least one of the plurality of heating devices (220) based on the surface temperature of the electrode. For example, if the temperature increase rate of the surface temperature of the electrode in the last drying zone exceeds a threshold value, the processor (130) may reduce the output of the heating device arranged in the last drying zone.

[0094] According to one embodiment, the processor (130) can control the drying device (200) differently depending on the type of electrode.

[0095] According to one embodiment, the processor (130) may control the output of all heating devices when the electrode is a cathode. Furthermore, the processor (130) may control the output of a heating device corresponding to a drying zone from which the electrode is discharged among the plurality of heating devices when the electrode is an anode. In other words, the output of a heating device provided in the last drying zone may be controlled. For example, when the electrode is an anode, the processor (130) may set the output of a heating device corresponding to a drying zone other than the last drying zone to 0.

[0096] For example, the processor (130) can obtain outputs of multiple heating devices (220) from the same deep learning model and selectively apply at least a portion of the outputs of the multiple heating devices (220) depending on the type of electrode.

[0097] In another example, the processor (130) may generate different deep learning models with different input variable learning applied depending on the type of electrode, and may apply the output of the heating device output from the corresponding deep learning model depending on the type of electrode. For example, if the electrode is an anode, the deep learning model may be generated and trained by applying the input variables corresponding to the last drying zone and excluding the input variables corresponding to other drying zones.

[0098] FIG. 2 is a drawing showing the configuration of a drying device (200) according to one embodiment disclosed in this document.

[0099] Referring to FIG. 2, the drying device (200) may include a plurality of drying zones (211, 213, 215). The electrode may move in a predetermined direction (D) within the drying device (200).

[0100] In addition, the drying device (200) may include a plurality of heating devices (221, 223, 225) arranged corresponding to each drying zone. For example, a first heating device (221) may be arranged in a first drying zone (211), a second heating device (223) may be arranged in a second drying zone (213), and a third heating device (225) may be arranged in a third drying zone (215).

[0101] Each of the plurality of heating devices (221, 223, 225) may be arranged on the upper and lower surfaces with respect to the transport path (20) through which the electrodes are transported. For example, the heating device (221) of the first drying zone (211) may be configured as a pair of heating devices, one arranged on the upper surface and one arranged on the lower surface with respect to the transport path (20).

[0102] The electrode can move inside the drying device (200) in a predetermined direction (D), and the plurality of heating devices (221, 223, 225) of the drying device (200) can emit heat to dry the electrode. For example, the drying device (200) can dry the electrode by applying heat to the electrode surface to remove moisture from the electrode for which the notching process has been completed.

[0103] FIG. 3 is a diagram showing an example of the structure of a deep learning model according to one embodiment disclosed in this document.

[0104] Referring to FIG. 3, an example of the structure of a neural network is illustrated when the deep learning model is an LSTM model. According to one embodiment, the LSTM model may include an input layer (310), a concatenation layer (320), and a hidden layer (330).

[0105] In an LSTM model, a sequence length parameter for input data can be defined. Here, the sequence length can be a parameter indicating the structure of the data, such as the length of the input data input to the LSTM model. In one embodiment, the sequence length of the LSTM model can be set to the number of input variables of the input data. For example, if six different input variables are input to the LSTM model, the sequence length can be set to 6.

[0106] According to one embodiment, the processor (130) may apply different deep learning models depending on the type of electrode, and may generate different deep learning models for this purpose. In this case, the number of input variables used for training the deep learning model may vary depending on the type of electrode, and accordingly, the sequence length may be set differently for each.

[0107] According to one embodiment, among the input variables of the deep learning model, the target temperature of the electrode surface at the target time point may be input to the hidden layer (330). In this case, the sequence length of the deep learning model may be set to a value obtained by subtracting the number of input variables corresponding to the target temperature of the electrode surface at the target time point from the number of input variables.

[0108] Figure 3 illustrates an example of a case where the sequence length is 4, and the circular dots represent each node. Figure 3 illustrates an example of a case where the sequence length is 4, but is not limited thereto.

[0109] Input data corresponding to the sequence length of the LSTM model can be input to the input layer (310). The input data input to the input layer (310) can pass through a plurality of corresponding sequence nodes according to the input location. For example, a plurality of sub-neural networks (311, 313, 315, 317) including a plurality of sequence nodes for each input location can be included. The plurality of sub-neural networks (311, 313, 315, 317) can learn the relationship between nodes.

[0110] The output of the last nodes (319) through which each input data passes can pass through a combination layer (320). The combination layer (320) can combine the results of each input data passing through the input layer (310).

[0111] The output passing through the coupling layer (320) may pass through the hidden layer (330). At this time, the hidden layer (330) may include a number of nodes corresponding to the number of drying zones. For example, if the drying device (200) includes a first drying zone, a second drying zone, and a third drying zone, the hidden layer (330) may include three nodes, as illustrated in FIG. 3.

[0112] Each node of the hidden layer (330) may be input with data regarding the target temperature of the electrode surface at the target time point of each drying zone. For example, as in the example described above, each node of the hidden layer (330) may be input with data regarding the target temperature of the electrode surface at the first target time point (tout(9)) and the target temperature of the electrode surface at the second target time point (tout(15)).

[0113] That is, data on the output of the bonding layer (320) and the target temperature of the electrode surface at the target time point of each drying zone can be input into the hidden layer (330), and the hidden layer (330) can ultimately derive the output of each of the plurality of drying zones (220) as output values ​​(yin, ymid, yout).

[0114] FIG. 4 is a diagram illustrating an example of the results of applying a deep learning model according to one embodiment disclosed in this document. FIG. 4 illustrates the results of achieving the target temperature of an electrode during a drying process using a drying device (200). A higher rate of achieving the target temperature of the electrode may result in a lower defect rate during the electrode drying process, indicating better performance of the trained deep learning model. Here, the target temperature of the electrode may refer to the target temperature at the aforementioned target time point.

[0115] For example, referring to FIG. 4, when the moving speed of the electrode is 1000 mm / s, it can be confirmed that the target temperature achievement rate of the electrode is 98.8% for the cathode and 92.8% for the anode as a result of controlling the output of multiple heating devices (220) using the learned deep learning model.

[0116] The target temperature achievement rate of the electrode can be calculated as the ratio of the number of times the electrode temperature reaches the target temperature among the number of times the drying device (200) is driven after floating. For example, in the case of the cathode, the target temperature is reached 238 times out of a total of 241 times, so the target temperature achievement rate can be calculated as 98.8%, and in the case of the anode, the target temperature is reached 90 times out of a total of 97 times, so the target temperature achievement rate can be calculated as 92.8%.

[0117] In this way, the electronic device (100) can control the output of multiple heating devices (220) by using the learned deep learning model when the drying device (200) is operated after floating, thereby improving the drying performance of the electrode and reducing the failure rate of the electrode.

[0118] FIGS. 5 to 8 are diagrams showing examples of correlation analysis for determining input variables of a deep learning model according to one embodiment disclosed in this document.

[0119] First, referring to FIGS. 5a and 5b, graphs are shown showing a one-to-one correlation between various variables and the target temperature of the discharge zone electrode surface when the electrode moving speeds are different, 1000 mm / s (FIG. 5a) and 300 mm / s (FIG. 5b), respectively.

[0120] The graphs (511 to 517) of Fig. 5a are scatter plots showing the relationship between the discharge zone electrode surface temperature and log_dead time, log_previous operating time, initial temperature of the input zone electrode surface, initial temperature of the output zone electrode surface, input zone electrode surface temperature at the target time, output of the input zone heating device, and output of the discharge zone heating device, respectively, when the moving speed of the electrode is 1000 mm / s. Similarly, the graphs (521 to 527) of Fig. 5b show the results when the moving speed of the electrode is 300 mm / s.

[0121] Referring to FIGS. 5a and 5b, it can be confirmed that the relationship between each input variable and the target temperature of the discharge zone electrode surface is similar at different electrode velocities (1000 mm / s and 300 mm / s). For example, comparing graphs (513) and (523) confirms that the scatter plots have similar trends.

[0122] That is, it can be inferred that the input variables illustrated in FIGS. 5a and 5b have a significant relationship with the target temperature of the discharge zone electrode surface even at different electrode movement speeds.

[0123] In order to analyze the relationship more specifically, the correlation values ​​with the output characteristics can be confirmed for the input variables shown in FIGS. 5a and 5b, and the results of analyzing the correlation values ​​for the input variables are shown in FIG. 6.

[0124] Table (610) of Fig. 6 shows the results of correlation analysis between input variables and the target temperature of the discharge zone electrode surface, table (620) shows the results of correlation analysis between input variables and the output of the input zone heating device, and table (630) shows the results of correlation analysis between input variables and the output of the discharge zone heating device.

[0125] Referring to Table (610) to Table (630) of Fig. 6, it can be confirmed that the correlation between each input variable and the Y value at different line speeds is similar, and that the correlation value has a statistically significant value.

[0126] However, among the multiple input variables, the variables of log_previous operating time and target point input zone electrode surface temperature were found to have no correlation with the output of the input zone heating device at 1000 mm / s, and the variables of log_previous operating time were found to have no correlation with the output of the discharge zone heating device at 1000 mm / s.

[0127] However, even if a specific input variable does not have a one-to-one correlation with a specific output characteristic, there may be interactions between input variables, so correlation analysis that considers interactions between input variables is necessary.

[0128] To this end, the results of analyzing the correlation by considering the interaction between input variables are shown in Figure 7.

[0129] Referring to the table illustrated in Figure 7, the results of analyzing the correlation between two randomly selected variables among multiple input variables and the output characteristics are depicted as each element of the matrix. More specifically, the value of each element can represent the frequency with which the two variables are judged to have a high correlation with the output characteristics when selected. For example, a correlation value of 0.7 or higher considering the interaction indicates a strong correlation, and therefore, the number of times the correlation value is 0.7 or higher can be counted.

[0130] In Fig. 7, it can be confirmed that the important variables showing correlation with the output characteristics are similar in both cases where the line speed is 1000 mm / s and 300 mm / s. That is, when checking the frequency of variable usage corresponding to each component, it can be said that the higher the frequency of variable usage, the higher the correlation with the output characteristics, and it can be confirmed that the frequency distribution of the input variables was derived similarly in the cases where the line speed is 1000 mm / s and 300 mm / s.

[0131] Correlation analysis, which considers the interactions between input variables, can account for these interactions, but it is difficult to analyze the simultaneous impact on multiple output characteristics. Therefore, a multivariate analysis of variance (MANOVA) can be performed between the input variables and output characteristics. The results of the MANOVA are shown in Figure 8.

[0132] Referring to Fig. 8, the influence of input variables on the output characteristics at different electrode movement speeds can be confirmed. Fig. 8 illustrates the influence of input variables on the output of the input zone heating device (SCR_IN) and the output of the exhaust zone heating device (SCR_OUT) among the output characteristics. More specifically, Fig. 8 shows the result values ​​according to various calculation methods derived as the Pr>F term. Here, the Pr>F term is a value representing a correlation factor, and if the value is less than 0.05, it can be evaluated that there is a significant correlation.

[0133] That is, referring to Fig. 8, the input variables analyzed in Figs. 5 to 8 can be evaluated as having a significant correlation with the output characteristics of the heating device, and as a result, the above input variables can be utilized for learning a deep learning model.

[0134] FIG. 9 is a flowchart for explaining a control method of an electronic device according to one embodiment disclosed in this document.

[0135] Referring to FIG. 9, at step S910, the processor (130) may obtain operating status information of the drying device (200) when the drying device (200) is operated after being floated. The operating status information of the drying device (200) may include, for example, the float time of the drying device, the previous operating time, and the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged. For example, the processor (130) may obtain such information from the drying device (200) through the communication circuit (110).

[0136] At step S920, the processor (130) may obtain status information of at least one electrode inserted into the drying device (200). The status information of the electrode may include, for example, an initial temperature of an electrode surface located in each of a plurality of drying zones and a target temperature of the electrode surface at a target time point. For example, the processor (130) may obtain such information from the drying device (200) via the communication circuit (110).

[0137] At step S930, the processor (130) can determine the output of each of the plurality of heating devices (220). For example, the processor (130) can input the operating status information of the drying device and the status information of at least one electrode into a learned deep learning model to confirm the output of each of the plurality of heating devices (220).

[0138] At step S940, the processor (130) can control the plurality of heating devices (220). For example, the processor (130) can control the output of each of the determined plurality of heating devices (220) as the initial output of each of the plurality of heating devices (220) when the drying device (200) is operated after floating.

[0139] Fig. 10 is a flowchart illustrating a specific control method of an electronic device according to one embodiment disclosed in this document. In Fig. 10, descriptions of content overlapping with Fig. 9 will be omitted.

[0140] Referring to FIG. 10, in step S1030, the processor (130) may input operating status information of the drying device (200) and status information of at least one electrode into the learned deep learning model. For example, the processor (130) may input data such as the idle time and previous operating time of the drying device as operating status information of the drying device (200), and data such as the initial temperature of the electrode surface located in each of a plurality of drying zones as status information of the electrode into the learned deep learning model.

[0141] At step S1040, the processor (130) can determine the output of each of the plurality of heating devices (220). The processor (130) can check the output of each of the plurality of heating devices (220) output from the deep learning model and determine this as the output of each of the plurality of heating devices (220) when the drying device (220) is operated after floating.

[0142] The processor (130) can control multiple heating devices (220) depending on the type of electrode. If the electrode is a cathode, the processor (130) can proceed to step S1050, and if the electrode is an anode, the processor (130) can proceed to step S1060.

[0143] At step S1050, the processor (130) can control the output of all heating devices when the electrode is a cathode. The processor (130) can control the output identified from the deep learning model as the initial output of each heating device.

[0144] At step S1060, the processor (130) can control the output of the heating device corresponding to the drying zone from which the electrode is discharged if the electrode is positive. That is, the processor (130) can control the output of the heating device of the discharge zone. For example, the processor (130) can set the output of the heating devices located in other drying zones, excluding the heating device of the discharge zone, to 0.

[0145] FIG. 11 is a flowchart illustrating a method for acquiring a trained deep learning model according to an embodiment disclosed in this document. In FIG. 11, the description will assume that the training of the deep learning model is performed on an electronic device (100). However, as described above, the creation and training of the deep learning model can of course be performed on a separate device (e.g., an external server) other than the electronic device (100).

[0146] Referring to FIG. 11, in step S1110, the processor (130) may select a plurality of variables. More specifically, the processor (130) may select a plurality of variables related to at least one of the driving state of the drying device (200) and the state of the electrode.

[0147] At step S1120, the processor (130) may determine input variables of the deep learning model from among a plurality of variables. For example, the processor (130) may analyze the correlation between the plurality of variables and the target temperature of the electrode surface according to the movement speed of the electrode. Furthermore, based on the analysis results, the processor (130) may determine at least some of the plurality of variables as input variables of the deep learning model.

[0148] At step S1130, the processor (130) may configure a learning input data set and a learning output data set for training a deep learning model. For example, the processor (130) may configure learning input data composed of data for determined input variables.

[0149] At step S1140, the processor (130) can input a learning input data set and a learning output data set into the deep learning model. That is, the processor (130) can input a learning input data set and a learning output data set into the deep learning model, thereby repeatedly training the deep learning model.

[0150] At step S1150, the processor (130) may acquire a trained deep learning model. As a result of repeated training of the deep learning model, the processor (130) may acquire a trained deep learning model optimized for target characteristics. For example, the processor (130) may acquire a trained deep learning model optimized for deriving the output of each of the multiple heating devices (220) of the drying device (200).

[0151]

[0152] Meanwhile, the present specification and drawings disclose preferred embodiments of the present disclosure, and although specific terms are used, they are used in a general sense only to easily explain the technical contents of the present disclosure and to help understand the embodiments, and are not intended to limit the scope of the present disclosure. It will be apparent to those skilled in the art to which the embodiments of the present disclosure pertain that other modified examples based on the technical idea of ​​the present disclosure are possible in addition to the embodiments disclosed herein.

[0153] The device or terminal according to the above-described embodiments may include a processor, a memory for storing and executing program data, a permanent storage such as a disk drive, a communication port for communicating with an external device, a user object device such as a touch panel, a key, a button, etc. The methods implemented as software modules or algorithms may be stored on a computer-readable recording medium as computer-readable codes or program commands that can be executed on the processor. Here, the computer-readable recording medium includes a magnetic storage medium (e.g., a read-only memory (ROM), a random-access memory (RAM), a floppy disk, a hard disk, etc.) and an optical reading medium (e.g., a CD-ROM, a Digital Versatile Disc (DVD)). The computer-readable recording medium may be distributed to computer systems connected through a network, so that the computer-readable code can be stored and executed in a distributed manner. The medium is readable by a computer, stored in a memory, and executed by a processor.

[0154] The present embodiment may be represented by functional block configurations and various processing steps. These functional blocks may be implemented by various hardware and / or software configurations that perform specific functions. For example, the embodiment may employ integrated circuit configurations such as memory, processing, logic, look-up tables, etc., which may perform various functions under the control of one or more microprocessors or other control devices. Similarly, the present embodiment may be implemented in a programming or scripting language such as C, C++, Java, assembler, Python, etc., including various algorithms implemented as a combination of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms that execute on one or more processors. Furthermore, the present embodiment may employ conventional techniques for electronic configuration, signal processing, and / or data processing. Terms such as "mechanism," "element," "means," and "composition" can be used broadly and are not limited to mechanical or physical structures. These terms can also encompass a series of software routines, such as those associated with a processor.

Claims

1. In electronic devices, A communication circuit communicating with a drying device including a plurality of drying zones for drying the inserted electrodes and a plurality of heating devices arranged to correspond to each of the plurality of drying zones and providing heat to the electrodes; Memory that stores one or more instructions; and Contains a processor, The one or more instructions, when executed, cause the processor to: When the above drying device is operated after floating, the operating status information of the above drying device is obtained, Obtaining status information of at least one electrode inserted into the drying device, Based on the driving status information of the drying device and the status information of the at least one electrode, the output of each of the plurality of heating devices is determined, configured to control the plurality of heating devices based on the output of each of the plurality of heating devices determined above, Electronic devices.

2. In paragraph 1, The operating status information of the above drying device is: Including the dead time of the drying device, the previous operating time and the initial temperature of the heating device corresponding to the drying zone from which the electrode is discharged, The status information of the above electrode is: Including the initial temperature of the electrode surface located in each of the plurality of drying zones and the target temperature of the electrode surface at the target time, Electronic devices.

3. In paragraph 2, The above target time is, Including the time elapsed from the time the electrode is inserted into the drying device until the inserted electrode reaches the drying zone from which the electrode is discharged among the plurality of drying zones. Electronic devices.

4. In paragraph 1, The above processor, Inputting the operating status information of the above drying device and the status information of at least one electrode into the learned deep learning model, configured to check the output of each of the plurality of heating devices output from the deep learning model, Electronic devices.

5. In paragraph 4, The above deep learning model A plurality of driving status information of the drying device corresponding to a plurality of driving of the drying device and a plurality of status information of a plurality of electrodes are used as a learning input data set, The output of each of the plurality of heating devices corresponding to the plurality of drives of the drying device is used as a learning output data set. A learning model that models the correlation between the above learning input data set and the above learning output data set, Electronic devices.

6. In paragraph 4, The above-mentioned trained deep learning model includes an LSTM model. Electronic devices.

7. In paragraph 4, The above processor, Selecting a plurality of variables related to at least one of the driving state of the drying device and the state of the electrode, By analyzing the correlation between the plurality of variables and the target temperature of the electrode surface according to the movement speed of the electrode, the input variables of the deep learning model are determined. Electronic devices.

8. In paragraph 1, The above processor, If the above electrode is a cathode, it controls the output of all heating devices, If the electrode is an anode, the output of the heating device corresponding to the drying zone from which the electrode is discharged among the plurality of heating devices is controlled. Electronic devices.

9. In paragraph 1, The above multiple dry zones are: Comprising a first drying zone, a second drying zone, and a third drying zone arranged along the movement path of the electrode, Electronic devices.

10. An electronic device comprising a plurality of drying zones for drying an inserted electrode, and controlling a plurality of heating devices arranged to correspond to each of the plurality of drying zones and providing heat to the electrode, When the drying device is operated after floating, a step of obtaining operating status information of the drying device; A step of obtaining status information of at least one electrode inserted into the drying device; A step of determining the output of each of the plurality of heating devices based on the driving status information of the drying device and the status information of the at least one electrode; and A step of controlling the plurality of heating devices based on the output of each of the plurality of heating devices determined above, Control method.

11. A non-transitory computer-readable recording medium having recorded thereon a program for executing the method of Article 10 on a computer.

Images