Rolling apparatus for manufacturing electrodes and method for operating the rolling apparatus

Abstract

The present disclosure relates to a rolling apparatus for manufacturing an electrode and a method of operating the same. A mechanism is provided for inputting a rolling gap between a first roller and a second roller for rolling an electrode, thereby minimizing quality deviation of the electrode. To this end, the present disclosure provides a configuration for predicting a rolling gap value using a gap prediction model and performing control such that a rolling gap between a first roller and a second roller becomes the predicted rolling gap value.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority and benefit to Korean Patent Application No. 10-2024-0138764, filed on October 11, 2024, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] This disclosure relates to a rolling mill for manufacturing electrodes and a method of operating the rolling mill, wherein the rolling gap is controlled during the rolling process. Background Technology

[0004] Commercially available rechargeable batteries include nickel-cadmium (NiCd), nickel-metal hydride (NiMH), nickel-zinc (NiZn), and lithium-ion batteries. Among these types of batteries, lithium-ion batteries have advantages because, compared to nickel-based batteries, they are easy to charge and discharge due to the almost complete absence of a memory effect, have a very low self-discharge rate, and possess high energy density.

[0005] The manufacturing process of lithium-ion batteries mainly consists of three stages: electrode forming, assembly, and formation. Electrode forming includes active material mixing, electrode coating, rolling, slitting, and winding. Among these processes, rolling is the process of pressing the electrode to the desired thickness by passing it between two rollers. Rolling reduces the thickness of the coated electrode to increase capacity density and improve the adhesion between the current collector and the active material.

[0006] In the rolling process, the electrode thickness is determined by the rolling gap between the first and second rolls. The lower roll (the second roll) uses a servo valve for position adjustment and rolling. However, conventionally, the rolling gap is manually adjusted based on the engineer's experience. This manual adjustment leads to variations in the quality of the rolled electrode due to differences in the engineer's experience.

[0007] The information disclosed in this section is intended to enhance understanding of the background of this disclosure, and therefore may contain information that does not constitute related (or prior art). Summary of the Invention

[0008] This disclosure aims to provide a rolling mill for manufacturing electrodes and a method of operating the rolling mill, wherein a rolling gap value between a first roller and a second roller is automatically input during the rolling of the electrode, thereby minimizing deviations in the rolled electrode.

[0009] However, the purposes achieved by this disclosure are not limited to those described above, and other purposes not described will be clearly understood by those skilled in the art from the following description.

[0010] According to one aspect of this disclosure, a rolling apparatus for manufacturing electrodes is provided, the rolling apparatus comprising: a rolling unit including a first roller and a second roller configured to form a rolling gap through which an electrode passes between the first roller and the second roller; a roller position adjustment unit configured to adjust the position of the second roller; and a control device configured to predict a rolling gap value using a gap prediction model and to control the roller position adjustment unit such that the rolling gap becomes the predicted rolling gap value.

[0011] The roller position adjustment unit may include: a drive unit connected to the second roller; and a control valve configured to receive a predicted roller gap value from a control device and control the drive unit to adjust the position of the second roller such that the roller gap becomes the predicted roller gap value.

[0012] When a conditional inference command is input through the user interface, the control device can input the downtime and previous gap value into the gap prediction model to predict the roll gap value.

[0013] Downtime can be represented as the difference between the start time of the current roll pressing process and the end time of the previous roll pressing process.

[0014] The rolling equipment may further include a thickness sensor configured to measure the thickness of an electrode passing between the first and second rollers.

[0015] The control device can compare the thickness of the electrode measured by the thickness sensor with a preset reference thickness range, and can adjust the roller gap when the thickness of the electrode exceeds the reference thickness range.

[0016] The rolling mill may further include: a server configured to collect process-related data, including at least one of standard data, derived data, and process data for a past set time, and the server may be configured to generate a gap prediction model using the collected process-related data.

[0017] By using process-related data as input characteristics and electrode thickness as output characteristics, the server can select the necessary factors for modeling by analyzing the correlation between input and output characteristics. The selected factors can be applied to each of multiple machine learning models to predict the roll gap value. The performance of each machine learning model can be evaluated based on the predicted roll gap value. One of the machine learning models can be selected based on the performance of each machine learning model, and the selected machine learning model can be provided as the gap prediction model.

[0018] Servers can use feature importance techniques to analyze the correlation between input and output characteristics.

[0019] The server can use an exponential function-based regression equation to generate a gap prediction model, which takes selected factors as input and outputs the roll gap value.

[0020] According to one aspect of the invention, a method of operating a rolling mill for manufacturing electrodes is provided, the rolling mill comprising: a first roller and a second roller forming a rolling gap through which an electrode passes; and a control device for controlling a roller position adjustment unit for adjusting the position of the second roller, the method comprising: using a gap prediction model to predict a rolling gap value, and controlling the roller position adjustment unit to adjust the position of the second roller such that the rolling gap becomes the predicted rolling gap value.

[0021] However, the effects achievable through this disclosure are not limited to those described above, and those skilled in the art can clearly understand from the detailed description other effects not described. Attached Figure Description

[0022] The accompanying drawings illustrate various embodiments of the present disclosure, and together with the detailed description, aspects and features of the present disclosure are described. The present disclosure is not limited to the embodiments depicted in the drawings, in which:

[0023] Figure 1 and Figure 2 This is a conceptual diagram of a rolling mill apparatus for manufacturing electrodes according to an embodiment of the present disclosure;

[0024] Figure 3 This is a view of the second roller position adjustment unit according to an embodiment of the present disclosure;

[0025] Figure 4 This is used to describe what happens when the rolling mill stops according to embodiments of the present disclosure.

[0026] Figure 5 This is a schematic block diagram illustrating the configuration of a control device according to an embodiment of the present disclosure;

[0027] Figure 6 This is an example diagram used to describe the display screen of the roll gap according to embodiments of the present disclosure;

[0028] Figure 7 This is a schematic block diagram illustrating the configuration of a server according to an embodiment of the present disclosure;

[0029] Figure 8 It is an exemplary chart used to describe the factors necessary for modeling a gap prediction model according to embodiments of the present disclosure;

[0030] Figure 9 An exemplary graph is shown to describe the gap adjustment deviation according to embodiments of the present disclosure;

[0031] Figure 10 This is a flowchart of a method for operating a rolling mill for manufacturing electrodes according to an embodiment of the present disclosure;

[0032] Figure 11 An exemplary graph is shown to describe the thickness trend of an electrode according to embodiments of the present disclosure; and

[0033] Figure 12 This is a flowchart of a method for generating a gap prediction model according to an embodiment of the present disclosure. Detailed Implementation

[0034] In the following description, various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The terms or words used in this specification and claims should not be construed as limited to their ordinary or dictionary meaning, but should be interpreted as consistent with the technical concept of the present disclosure, based on the principle that the inventor is capable of being his / her own lexicographer to appropriately define the concepts of the terms in order to best interpret his / her invention.

[0035] The embodiments described in this specification and the constructions shown in the accompanying drawings are merely some of the embodiments of this disclosure and do not represent all the technical concepts, aspects, and features of this disclosure. Accordingly, it should be understood that various equivalents and modifications that can replace or modify the embodiments described herein can exist.

[0036] It will be understood that when a component or layer is referred to as being "on" another component or layer, "connected to," or "coupled to" another component or layer, it can be directly on, directly connected to, or coupled to that other component or layer, or there may be one or more intermediary components or layers. However, when a component or layer is referred to as being "directly on" another component or layer, "directly connected to," or "directly coupled to" another component or layer, there are no intermediary components or layers. For example, when a first component is described as being "coupled" or "connected" to a second component, the first component can be directly coupled to or connected to the second component, or the first component can be indirectly coupled to or connected to the second component via one or more intermediary components.

[0037] In the accompanying drawings, the dimensions of various elements, layers, etc., may be exaggerated for clarity. The same reference numerals designate the same elements. As used herein, the term "and / or" includes any and all combinations of one or more of the related listed items. Furthermore, the use of "may" in describing embodiments of this disclosure refers to "one or more embodiments of this disclosure." When expressions such as "at least one of..." and "any one of..." follow a list of elements, they modify the entire list of elements and not individual elements within that list. When a list of elements A, B, and C is specified using terms such as "at least one of A, B, and C," "at least one selected from the group of A, B, and C," or "at least one selected from A, B, and C," the term may refer to any and all appropriate combinations or subsets of A, B, and C, such as A, B, C, A and B, A and C, B and C, or A and B and C. As used herein, the terms "use," "used," and "used" may be considered synonymous with the terms "utilize," "exploited," and "exploited," respectively. As used herein, the terms “substantially,” “about,” and similar terms are used as approximations and not as terms of degree, and are intended to explain the inherent variations in measurements or calculations that will be recognized by one of ordinary skill in the art.

[0038] It will be understood that although the terms first, second, third, etc., may be used herein to describe various elements, components, areas, layers, and / or portions, these elements, components, areas, layers, and / or portions should not be limited by these terms. These terms are used to distinguish one element, component, area, layer, or portion from another element, component, area, layer, or portion. Therefore, the first element, component, area, layer, or portion discussed below may be referred to as the second element, component, area, layer, or portion without departing from the teachings of the exemplary embodiments.

[0039] For ease of description, spatial relative terms such as “below,” “under,” “below,” “above,” and “over” are used herein to describe the relationship between one element or feature as illustrated in the figures and another element(s). It will be understood that, in addition to the orientation depicted in the figures, the spatial relative terms are intended to cover different orientations of the device in use or operation. For example, if the device in the figures is flipped, an element described as “below” or “under” other elements or features will subsequently be oriented “above” or “over” that other element or feature. Thus, the term “below” can encompass both above and below orientations. The device may be oriented in other ways (rotated 90 degrees or in other orientations), and the spatial relative descriptors used herein should be interpreted accordingly.

[0040] The terminology used herein is for the purpose of describing various embodiments of this disclosure and is not intended to limit this disclosure. As used herein, the singular forms “a” and “the (described)” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that, when used in this specification, the terms “comprising,” “including,” “containing,” and / or variations thereof specify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0041] Furthermore, any numerical range disclosed and / or enumerated herein includes all subranges of the same numerical precision falling into the enumerated range. For example, the range “1.0 to 10.0” includes all subranges between (and including) the enumerated minimum value of 1.0 and the enumerated maximum value of 10.0, that is, a minimum value equal to or greater than 1.0 and a maximum value equal to or less than 10.0, such as, for example, 2.4 to 7.6. Any maximum numerical limit enumerated herein includes all lower numerical limits falling into it, and any minimum numerical limit enumerated herein includes all higher numerical limits falling into it. Accordingly, the applicant reserves the right to amend this specification (including the claims) to expressly enumerate any subranges falling into the ranges expressly enumerated herein. All such ranges are intended to be inherently described in this specification such that any modification to any of these expressly enumerated subranges will comply with the requirements of the Chinese Patent Law.

[0042] Referring to two compared elements, features, etc., as "identical" can mean that they are "substantially identical." Therefore, the term "substantially identical" can include cases where there is a relatively low deviation (e.g., 5% or less) in the art. Furthermore, when a parameter is said to be uniform in a given region, this can mean that it is uniform in terms of its mean.

[0043] Throughout this specification, unless otherwise stated, each element may be singular or plural.

[0044] When an element is described as being set (or positioned or placed) "above (or below)" or "on top (or below)" a component, this can mean that the element is positioned to contact the upper (or lower) surface of the component, and it can also mean that another component may be located between the component and any element set (or positioned or placed) on (or below) the component.

[0045] Furthermore, it should be understood that when one element is referred to as "coupled," "linked," or "connected" to another element, these elements may be directly "coupled," "linked," or "connected" to each other, or there may be an intervening element between them through which the element can be "coupled," "linked," or "connected" to the other element. Additionally, when one part is referred to as "electrically coupled" to another part, that part may be directly connected to the other part, or there may be an intervening part between them, such that the part and the other part are indirectly connected to each other.

[0046] Throughout this specification, unless otherwise stated, when “A and / or B” is stated, it means A, B, or A and B. That is, “and / or” includes any or all of the enumerated items. Unless otherwise specified, when “C to D” is stated, it means above C and below D.

[0047] Figure 1 and Figure 2 This is a conceptual diagram of a rolling mill apparatus for manufacturing electrodes according to an embodiment of the present disclosure. Figure 3 This is a view of the second roller position adjustment unit according to an embodiment of the present disclosure. Figure 4 This is an exemplary graph used to describe the change in the position of the second roller when the roller pressing equipment stops, according to an embodiment of the present disclosure.

[0048] refer to Figure 1 and Figure 2 According to embodiments of the present disclosure, a rolling device for manufacturing electrodes may include one or more pairs of rollers 110, 120, 160 and 170, a second roller position adjustment unit 130, 140 and 150, a thickness sensor 180 and a control device 200.

[0049] exist Figure 1 In the diagram, the roller pressing device is illustrated as including a pair of rollers 110 and 120. The first roller 110 presses the electrode 10 from above, and the second roller 120 presses the electrode 10 from below.

[0050] exist Figure 2In the diagram, the rolling mill is illustrated as comprising two pairs of rollers: 110, 120, 160, and 170. Rollers positioned at the lower level can be defined as first-stage rollers 110 and 120, and rollers positioned at the upper level can be defined as second-stage rollers 160 and 170. The first-stage rollers 110 and 120 and the second-stage rollers 160 and 170 may include first rollers 110 and 160 pressing the electrode 10 from above and second rollers 120 and 170 pressing the electrode 10 from below. When the electrode 10 is placed into the rolling mill, the first-stage rollers 110 and 120 and the second-stage rollers 160 and 170 can each rotate to roll the electrode 10. Because the rolling mill comprises two pairs of rollers 110, 120, 160, and 170, the rolling process occurs twice. Because the first-stage rollers 110 and 120 and the second-stage rollers 160 and 170 perform the same process as... Figure 1 The pair of rollers 110 and 120 shown operate identically, therefore, for convenience, reference will be made to... Figure 1 The following description will be provided.

[0051] The electrode 10, supplied from the unwinder 20, is rolled as it passes between the first roller 110 and the second roller 120, and after passing through the conveyor 190, it is wound by the winder 30, thereby completing the manufacturing of the electrode 10. Here, the electrode 10 may be a sheet electrode (or an electrode plate), and may be a secondary battery electrode in which an electrode assembly coating layer comprising an electrode active material, a bonding agent, a conductive material, and a filler is formed on one or both surfaces of the current collector.

[0052] The first roller 110 and the second roller 120 face each other, and the positioning of the second roller 120 defines the roll gap through which the electrode 10 passes between the second roller 120 and the first roller 110. Here, the first roller 110 may be a roller positioned on the upper side, and the second roller 120 may be a roller positioned below the first roller 110. The first roller 110 and the second roller 120 are arranged vertically to provide the roll gap.

[0053] The first roller 110 may be formed in the shape of a cylinder having an axis in the longitudinal direction and may rotate about a first rotation axis. The second roller 120 may be provided facing the first roller 110, may be formed in the shape of a cylinder having an axis in the longitudinal direction and may rotate about a second rotation axis, while forming the electrode 10 through the roll gap through which it passes between the second roller 120 and the first roller 110.

[0054] The first roller 110 and the second roller 120 can be spaced apart from each other by a predetermined distance. The first roller 110 can be installed in a fixed position, and the second roller 120 can be installed in a movable position, so that the distance between the first roller 110 and the second roller 120 can be adjusted according to the thickness, material, etc. of the target to be rolled (i.e., electrode 10).

[0055] The second roller 120 can be in close contact with the electrode 10 and, together with the first roller 110, perform rolling to planarize the active material coated on the electrode 10. The first roller 110 and the second roller 120 can rotate at a specified speed to press and roll the electrode 10 through the rolling gap.

[0056] The second roller position adjustment units 130, 140, and 150 are components used to adjust the position of the second roller 120 and may include hydraulic cylinders. Specifically, as Figure 3 As shown, the second roller position adjustment units 130, 140 and 150 may include a drive unit 130, a position sensor 140 and a hydraulically operated control valve 150.

[0057] The drive unit 130 can be directly or indirectly connected to the second roller 120 to adjust the position of the second roller. The drive unit 130 can be a hydraulic cylinder that vertically moves a piston between various heights and can be controlled by the control device 200. The vertical movement of the second roller 120 can be achieved by adjusting the pressure of the hydraulic cylinder. Alternatively, the drive unit 130 can consist of a shaft or screw coupled to the second roller 120, and may further include an adjusting motor (not shown), and may further include gears for performing the rotational motion. However, the construction of the drive unit 130 and the method of adjusting the second roller 120 are not limited to these examples and can be applied to various configurations.

[0058] The drive unit 130 can adjust the rolling gap between the first roller 110 and the second roller 120 by adjusting the motor or the like. For example, the drive unit 130 can adjust the rolling gap by adjusting the position of the second roller 120 in the vertical direction (e.g., the Y-axis direction).

[0059] The position sensor 140 can sense the position of the second roller 120 or the drive unit 130, and can measure the position of the second roller 120 or the drive unit 130 based on a preset value according to the thickness of the electrode 10 coated with active material thereon. The position sensor 140 can be implemented in various ways, such as infrared method and scale method.

[0060] The control valve 150 can receive the position of the second roller 120 or the drive unit 130 from the position sensor 140. The control device 200 can control the control valve 150 to control the operation of the drive unit 130, and thereby adjust the position of the second roller 120.

[0061] The control valve 150 may include a pressure control valve (not shown) and a servo valve. Fluid generated from a hydraulic generator (not shown) passes through the pressure control valve, and the hydraulic pressure of the fluid allows the drive unit 130 to be raised or lowered by the servo valve. The position of the drive unit 130 can be sensed by a position sensor 140, and the control device 200, which receives the sensed position, controls the servo valve, thereby enabling real-time control of the position of the second roller 120.

[0062] Although only the second roller position adjustment units 130, 140, and 150 for the second roller 120 are illustrated and described in this embodiment, the present disclosure is not limited thereto. For example, the position adjustment unit may be connected only to the first roller 110. In another example, the position adjustment unit may be connected to both the first roller 110 and the second roller 120 to control the first roller 110 and the second roller 120, thereby adjusting the roller gap.

[0063] The first roller 110 or the second roller 120 may be connected to a connecting member (not shown) extending from each side thereof, and the connecting member may be coupled to a back pressure cylinder (not shown). The back pressure cylinder may have a hydraulic cylinder structure and may be vertically controlled to eliminate thickness deviations of the electrode 10 in the lateral direction (width direction) across the roller gap.

[0064] The thickness sensor 180 can measure the thickness of the electrode 10 as it passes between the first roller 110 and the second roller 120 and is pressed by the rollers. The thickness sensor 180 can transmit information related to the thickness of the electrode 10 being measured to the control device 200.

[0065] The thickness sensor 180 can calculate the thickness of the electrode 10 by changing the position of the second roller 120, either by directly detecting the thickness of the electrode 10 passing through the rolling mill, or by detecting the gap between the rollers. The thickness-related information obtained in this way can be transmitted to the control device 200. Figure 1 and Figure 2 An exemplary map shows the location of the thickness sensor 180, and this disclosure is not limited to the illustrated construction.

[0066] The rolling mill for manufacturing electrodes further includes a conveying component 190 for conveying the electrodes 10. The conveying component 190 may include, for example... Figure 1 and Figure 2 The multiple conveyor rollers shown. Figure 1 and Figure 2 An exemplary map illustrates the transport component 190, but this disclosure is not limited to what is shown.

[0067] The control device 200 can control the rolling gap through which the electrode 10 passes between the first roller 110 and the second roller 120. The control device 200 can control the rolling gap by controlling the position of the second roller 120 via a servo valve and a drive unit 130. When the electrode 10 is rolled using a rolling device configured as described above, the second roller 120, positioned below the first roller 110, is adjusted via a servo valve to roll the electrode 10, and the rolling amount can be determined based on the rolling gap value between the first roller 110 and the second roller 120.

[0068] When the rolling mill stops, the user interface 230 of the control device 200 (e.g., touch the pad) (see...) Figure 4 The first roller 110 outputs the same gap value, but the actual position of the second roller 120 may shift over time. That is, because the first roller 110 has a fixed structure, its position is fixed. However, because the second roller 120 has a vertically moving structure, the second roller 120 (i.e., the lower roller) moves as... Figure 4 As shown, it decreases over time. The decrease of the second roller 120 causes an increase in the roller gap. Accordingly, a gap correction amount is required for the roller pressing equipment.

[0069] Traditionally, the roll gap is manually adjusted based on the engineer's experience. However, this manual adjustment can lead to deviations in electrode quality, potentially resulting in shortened battery life and / or defects within the battery. Therefore, a technology is needed to minimize electrode quality deviations by automatically inputting the roll gap value between the first and second rolls during the rolling process.

[0070] According to this disclosure, quality deviations in the rolling process of electrode 10 are minimized by generating a gap prediction model that can automatically predict the roll gap value and using the generated gap prediction model to automatically predict the optimal roll gap value.

[0071] Therefore, the rolling equipment according to this disclosure may include a server 300 for collecting rolling process data, which uses the collected data to generate a gap prediction model. The server 300 may collect process-related data including at least one of rolling process data, standard data, and derived data, and the server 300 may use the collected process-related data to generate a gap prediction model.

[0072] Server 300 can distribute the gap prediction model to control device 200. Server 300 can be implemented as an edge server or a cloud server. (See reference...) Figure 7 Let's describe server 300 in detail.

[0073] In an embodiment, the control device 200 can use a gap prediction model to predict the roll gap value, and can use the predicted roll gap value to control the position of the second roller 120, thereby controlling the roll gap between the first roller 110 and the second roller 120. For example, the control device 200 can control the second roller position adjustment units 130, 140, and 150 to adjust the position of the second roller 120 so that the roll gap between the first roller 110 and the second roller 120 becomes the predicted roll gap value. (Refer to...) Figure 5 The control device 200 will be described in detail.

[0074] In the embodiments described above, server 300 generates a roll gap model, and control device 200 uses the gap prediction model to predict the roll gap value. However, in other embodiments, server 300, which generates the gap prediction model, and control device 200, which predicts the roll gap value, can also be implemented as a single device.

[0075] Figure 5 This is a schematic block diagram illustrating the configuration of a control device according to an embodiment of the present disclosure. Figure 6 This is an example diagram used to describe a display screen showing the roller gap according to an embodiment of the present disclosure.

[0076] refer to Figure 5 The control device 200 according to the embodiments of the present disclosure may include a communication module 210, a memory 220, a user interface 230, and a controller 240.

[0077] The communication module 210 provides a communication interface for transmitting or receiving signals in packet data between external devices (e.g., second roller position adjustment units 130, 140, and 150, servo valves, and server 300) in conjunction with a communication network. Furthermore, the communication module 210 can transmit position adjustment signals to the second roller position adjustment units 130, 140, and 150, and can receive information related to the thickness of electrode 10 from thickness sensor 180. Additionally, the communication module 210 may include the hardware and software necessary for transmitting or receiving signals such as control signals or data signals via wired or wireless connections with other network devices. The communication module 210 can be implemented in various forms, such as a short-range communication module, a wireless communication module, a mobile communication module, and a wired communication module.

[0078] The memory 220 stores multiple pieces of data related to the operation of the control device 200. Specifically, a gap prediction model that enables the prediction of the roll gap value can be stored in the memory 220, and multiple pieces of stored information can be selected by the controller 240 as needed. Furthermore, the memory 220 stores various types of data generated during the execution of an operating system or program (application or applet) used to drive the control device 200. In this context, the memory 220 is a general term encompassing both non-volatile storage devices that continuously maintain stored information even without power supply and volatile storage devices that require power to maintain stored information. Furthermore, the memory 220 can perform the function of temporarily or permanently storing data processed by the controller 240. Here, in addition to volatile storage devices that require power to maintain stored information, the memory 220 may also include magnetic storage media or flash memory storage media. However, this disclosure is not limited to these examples.

[0079] User interface 230 can receive instructions from the user or output results based on instructions from the user. User interface 230 can be implemented as, for example, a button, touch panel, touchpad, thin-film transistor liquid crystal display (TFT-LCD) panel, light-emitting diode (LED) panel, organic LED (OLED) panel, active matrix OLED (AMOLED) panel, flexible panel, etc.

[0080] When a user inputs a conditional inference command through user interface 230, user interface 230, under the control of controller 240, outputs a roll gap display screen including at least one of the following: downtime, predicted gap value, and previous gap value. Here, the predicted gap value can be a roll gap value predicted by a gap prediction model, and can include both work-side (W / S) and drive-side (D / S) predicted gap values. The previous gap value can be a roll gap value used during operation immediately preceding the current time, and can include both W / S and D / S previous gap values. Figure 6 As shown, the roll gap display screen can display the predicted gap value, including the W / S predicted gap value and the D / S predicted gap value, the previous gap value, including the W / S previous gap value and the D / S previous gap value, and the downtime. The user interface 230 can output both the previous gap value and the predicted gap value, allowing the user to easily check how much gap correction has been performed.

[0081] Controller 240 can be configured to control the overall operation of control device 200. For example, controller 240 can execute software (e.g., a program) stored in memory 220 to control components connected to controller 240 (e.g., at least one of communication module 210, memory 220, and user interface 230). Controller 240 can be implemented as an application-specific integrated circuit (ASIC), digital signal processor (DSP), programmable logic device (PLD), field-programmable gate array (FPGA), central processing unit (CPU), and / or microcontroller. However, this disclosure is not limited to these examples.

[0082] When a conditional inference command is input via user interface 230, controller 240 can predict the roll gap value by inputting the downtime and previous gap value into the gap prediction model. When the rolling equipment stops, the position of the second roll 120 may change over time, causing the roll gap to vary depending on the downtime of the rolling equipment. Therefore, the roll gap correction amount can be different, and the downtime can be an important variable in the gap prediction model. Since the previous gap value is the value required to calculate the roll gap correction amount, it can also be an important variable in the gap prediction model. Therefore, the gap prediction model can use downtime and the previous gap value as input variables and can output a predicted gap value.

[0083] Downtime can be the difference between the end point and the start point of the rolling process. In other words, when changing rolls between tasks, there are process end points (RollUp) and process start points (RollDown), and the downtime can be the difference between these two points. In other words, downtime can be the difference between the start time of the current rolling process and the end time of the previous rolling process.

[0084] The previous gap value can be the roll gap value at the end of the previous roll forming process. The previous gap value can include the W / S previous gap value and the D / S previous gap value. Here, W / S and D / S refer to the respective sides of the electrode 10 mounted on the unwinder 20. That is, for the roll forming process, the electrode 10 is divided into a certain number of sections in the length direction, where the section closest to the worker is defined as W / S, and the remaining sections are defined as D / S. For example, when a 1000m electrode 10 is divided into 10 equal parts, partition 1 can be in the range of 1m to 100m, partition 2 can be in the range of 101m to 200m, partition 3 can be in the range of 201m to 300m, partition 4 can be in the range of 301m to 400m, partition 5 can be in the range of 401m to 500m, partition 6 can be in the range of 501m to 600m, partition 7 can be in the range of 601m to 700m, partition 8 can be in the range of 701m to 800m, partition 9 can be in the range of 801m to 900m, and partition 10 can be in the range of 901m to 1000m. In this case, partitions 1 to 5 are W / S, and partitions 6 to 10 are D / S.

[0085] When input condition inference instructions are given, controller 240 can obtain the downtime of the rolling mill and the previous gap value of the rolling mill, and controller 240 can input the obtained downtime and the previous gap value into the gap prediction model. For example, controller 240 can obtain the difference between the start time of the current rolling process and the end time of the previous rolling process as the downtime.

[0086] When downtime and previous gap values ​​are input into the gap prediction model, controller 240 can predict the roll gap value. For convenience, the predicted roll gap value will be referred to as the predicted gap value in the following description. The predicted gap value may include both W / S predicted gap value and D / S predicted gap value.

[0087] When a two-stage roller press is used, the prior gap value can include the W / S prior gap value and D / S prior gap value of the first stage roller, as well as the W / S prior gap value and D / S prior gap value of the second stage roller. In this case, the downtime and prior gap value can be input into the gap prediction model, and the controller 240 can predict the predicted gap value, including the W / S predicted gap value and D / S predicted gap value of the first stage roller, as well as the W / S predicted gap value and D / S predicted gap value of the second stage roller.

[0088] After obtaining the predicted gap value, the controller 240 can control the position of the second roller 120. Specifically, the controller 240 can control the servo valves of the second roller position adjustment units 130, 140, and 150, so that the roller gap becomes the predicted gap value. That is, the controller 240 can control the servo valves and the drive unit 130 to adjust the position of the second roller 120, so that the second roller 120 moves to the position corresponding to the predicted gap value. In other words, the controller 240 can control the servo valves to drive the drive unit 130 according to the predicted gap value, thereby adjusting the position of the second roller 120 and controlling the roller gap between the first roller 110 and the second roller 110.

[0089] In this way, the controller 240 can control the electrode 10 through the roll gap it passes between the first roller 110 and the second roller 120. That is, the controller 240 can control the second roller position adjustment units 130, 140 and 150 to adjust the position of the second roller 120 according to the predicted gap value, thereby forming a roll gap corresponding to the predicted gap value.

[0090] After the position of the second roller 120 is adjusted and the rolling equipment is running, the thickness of the electrode 10 can be determined by the thickness sensor 180. When the thickness of the electrode 10 is within a preset reference thickness range, the controller 240 can operate to continuously execute the rolling process. When the measured thickness of the electrode 10 deviates from the reference thickness range, the controller 240 can additionally adjust the rolling gap. In this case, the rolling gap can be adjusted manually.

[0091] Figure 7 This is a schematic block diagram illustrating the configuration of a server according to an embodiment of the present disclosure. Figure 8 This is an exemplary chart used to describe the factors necessary for modeling a gap prediction model according to embodiments of this disclosure. Figure 9 An exemplary graph is shown to describe the gap adjustment deviation according to embodiments of the present disclosure.

[0092] refer to Figure 7 According to embodiments of the present disclosure, the server 300 may include a communication module 310, a memory 320, a database 330, and a processor 340.

[0093] The communication module 310 provides the necessary communication interface for transmitting or receiving signals in packet data between the server 300 and an external device (e.g., the control device 200) via a communication network. Furthermore, the communication module 310 can collect process-related data from the control device 200 and can send gap prediction models to the control device 200. The communication module 310 may include the hardware and software necessary for transmitting or receiving signals such as control signals or data signals via wired or wireless connections with other network devices. The communication module 310 can be implemented in various forms, such as a short-range communication module, a wireless communication module, a mobile communication module, and a wired communication module.

[0094] The memory 320 stores data related to the operation of the server 300. Specifically, the memory 320 may store programs (applications or applets) capable of collecting process-related data from multiple control devices 200, as well as programs (applications or applets) capable of using process-related data to generate gap prediction models. The processor 340 may select the information to be stored as needed. Furthermore, the memory 320 stores various types of data generated during the execution of the operating system or programs (applications or applets) used to drive the server 300. In this context, the memory 320 is a general term encompassing both non-volatile storage devices that continuously maintain stored information even when no power is supplied and volatile storage devices that require power to maintain stored information. Furthermore, the memory 320 may perform the function of temporarily or permanently storing data processed by the processor 340. Here, in addition to volatile storage devices that require power to maintain stored information, the memory 320 may also include magnetic storage media or flash memory storage media. However, this disclosure is not limited to these examples.

[0095] Database 330 can store process-related data collected through communication module 310. Here, process-related data may include at least one of process data, standard data, and derived data.

[0096] Processor 340 can be configured to control the overall operation of server 300. For example, processor 340 can execute software (e.g., a program) stored in memory 320 to control components connected to processor 340 (e.g., at least one of communication module 310, memory 320, and database 330). Processor 340 can be implemented as an ASIC, DSP, PLD, FPGA, CPU, and / or microcontroller. However, this disclosure is not limited to these examples.

[0097] Processor 340 can collect process-related data, including at least one of standard data, derived data, and process data for a past set time period. Processor 340 can use the collected process-related data to generate a gap prediction model.

[0098] The method by which processor 340 generates a roll gap prediction model will be described below.

[0099] The processor 340 can collect standard data, export data, and process data for a past set time from the control device 200. Here, the process data can be data from a rolling process for a past set time, and may include the thickness, temperature, speed, length, pressure, back pressure, load, tension, and roller speed of the electrode 10 (electrode plate). The standard data may include the center value (e.g., average value) of the electrode thickness, the upper limit of the electrode thickness, and the lower limit of the electrode thickness. The exported data can be data derived from the process data and the standard data, and may include, for example, downtime.

[0100] When process-related data, including at least one of process data, standard data, and derived data, is collected, the processor 340 can preprocess the process-related data. Preprocessing may include data merging, repetitive time processing, data sorting, missing value handling, and outlier handling.

[0101] Processor 340 can select the necessary factors for modeling by analyzing the correlation between preprocessed process-related data and electrode thickness. For example, by using the y-factor as electrode thickness and the x-factor as the remaining data after excluding electrode thickness from the process-related data, processor 340 can select a certain number of factors in order of their greatest impact on electrode thickness. In other words, by using electrode thickness as the output characteristic and the remaining data from the process-related data (excluding electrode thickness) as the input characteristic, processor 340 can select the necessary factors for modeling by analyzing the correlation between the input and output characteristics. Processor 340 can generate input characteristics in which the correlation between the input and output characteristics is greater than a preset value as a training dataset. In this case, processor 340 can use feature importance techniques to analyze the correlation between the input and output characteristics. Feature importance techniques can be measures that indicate how much influence a corresponding feature has on classifying a node when it branches in a decision tree.

[0102] For example, processor 340 can select factors from 302 rolling process factors through correlation analysis, such as Figure 8The 24 factors shown can include the first-stage roller position control value (W / S), the second-stage roller position control value (D / S), the electrode plate thickness of section 1 of the first-stage roller, the electrode plate thickness of section 2 of the first-stage roller, the electrode plate thickness of section 3 of the first-stage roller, the electrode plate thickness of section 4 of the first-stage roller, the electrode plate thickness of section 5 of the first-stage roller, the electrode plate thickness of section 6 of the first-stage roller, the electrode plate thickness of section 7 of the first-stage roller, and the electrode plate thickness of section 8 of the first-stage roller. Electrode plate thickness, electrode plate thickness of partition 9 of the first-stage roller, electrode plate thickness of partition 10 of the first-stage roller, electrode plate thickness of partition 1 of the second-stage roller, electrode plate thickness of partition 2 of the second-stage roller, electrode plate thickness of partition 3 of the second-stage roller, electrode plate thickness of partition 4 of the second-stage roller, electrode plate thickness of partition 5 of the second-stage roller, electrode plate thickness of partition 6 of the second-stage roller, electrode plate thickness of partition 7 of the second-stage roller, electrode plate thickness of partition 8 of the second-stage roller, electrode plate thickness of partition 9 of the second-stage roller, and electrode plate thickness of partition 10 of the second-stage roller.

[0103] Because the input variables of the gap prediction model include downtime and previous gap values, the downtime and previous gap values ​​can be derived using selected factors. The previous gap values ​​can be derived using the first-stage roll position control values ​​(W / S), the second-stage roll position control values ​​(D / S), and the third-stage roll position control values ​​(W / S and D / S). The downtime can be the interval between the time for measuring the final thickness in the first batch and the time for measuring the first thickness in the second batch. Therefore, the factor necessary for deriving the downtime can be the time for measuring the thickness of electrode 10, and the time for measuring the thickness of electrode 10 can be derived based on the electrode thickness after rolling. Accordingly, the downtime can be derived based on the plate thickness of sections 1 to 10 of the first-stage roll and the plate thickness of sections 1 to 10 of the second-stage roll.

[0104] The processor 340 can predict the roll gap value by applying selected factors to each of multiple machine learning models. Here, the machine learning models can be regression models. For example, machine learning models can include exponential regression, multiple linear regression (MLR), deep learning, genetic algorithms (GA), boosting trees, generative adversarial networks (GAN), artificial neural networks (ANN), ensembles, etc.

[0105] When predicting the roll gap value for each machine learning model, the processor 340 can evaluate the performance of each machine learning model based on the predicted roll gap value. That is, the processor 340 can use performance evaluation metrics based on the difference between the roll gap value predicted by each machine learning model and the actual roll gap value. Here, performance evaluation metrics may include mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), mean squared logarithmic error (MSLE), root mean squared logarithmic error (RMSLE), coefficient of determination (R² score), etc. Lower values ​​for MAE, MSE, RMSE, MSLE, and RMSLE indicate higher regression performance. A higher coefficient of determination (R² score) indicates higher regression performance.

[0106] When evaluating the performance of each machine learning model, processor 340 can select the best machine learning model based on the performance of each model. In this case, processor 340 can select the machine learning model with the best performance as the best machine learning model.

[0107] In a specific example, processor 340 can generate a gap prediction model using an exponential function-based regression equation that takes factors selected from process-related data over a past set time period (e.g., 3 to 6 months from the current time) as input and the roll gap value as output. In this case, processor 340 can generate an exponential regression model. As a result of analyzing process-related data, since the gap adjustment value (i.e., the difference between the predicted gap value and the previous gap value) does not increase indefinitely with increasing downtime, an exponential regression model can be used.

[0108] The range of parameters applied to the exponential regression equation can be specified by the user, and the optimal value of each parameter can be calculated using grid search techniques. Here, grid search can be a method of systematically trying various combinations of hyperparameters to optimize the model's performance. The processor 340 can perform parameter optimization by inputting the parameters derived through grid search into the exponential regression equation and calculating the error (MAE) between the predicted gap value and the actual gap value predicted by the exponential regression equation.

[0109] While the gap prediction model is learning, processor 340 can perform process capability index (Cpk) data preprocessing. Cpk data preprocessing can be performed during the learning process and can refer to calculating the initial thickness process capability for each batch within all batches during the learning period before performing learning, and subsequently removing batches with process capabilities less than or equal to a reference value (e.g., 1.5) for subsequent learning. Low process capability means high thickness deviation. When a batch with high thickness deviation is learned, it is removed because the probability of low prediction performance is high.

[0110] When performing Cpk data preprocessing, it can be confirmed that the gap adjustment bias in the gap prediction model is reduced, such as... Figure 9 As shown in the image.

[0111] Figure 10 This is a flowchart of a method for operating a rolling mill for manufacturing electrodes according to an embodiment of the present disclosure. Figure 11 An exemplary graph is shown to describe the thickness trend of an electrode according to an embodiment of the present disclosure.

[0112] refer to Figure 10 Server 300 collects process-related data, including standard data, exported data, and process data for a past set time (operation S1002). Here, process data may be data based on a rolling process for a past set time, and may include electrode plate thickness, temperature, speed, length, pressure, back pressure, load, tension, roller speed, etc. Standard data may include the center value (e.g., average value) of the electrode plate thickness, the upper limit of the electrode plate thickness, and the lower limit of the electrode plate thickness. Exported data may be data derived from process data and standard data, and may include, for example, downtime.

[0113] After performing operation S1002, server 300 uses process-related data to generate a gap prediction model (operation S1004). (Refer to...) Figure 12 This describes the method for generating a gap prediction model for server 300.

[0114] After performing operation S1004, server 300 transmits the gap prediction model to control device 200 (operation S1006), and control device 200 stores the gap prediction model (operation S1008).

[0115] After executing operation S1008, when the condition inference instruction is input (S1010), the control device 200 uses the gap prediction model to predict the roll gap value (operation S1012). Therefore, when the condition inference instruction is input, the control device 200 can obtain the downtime and previous gap value of the roll forming equipment, and can input the obtained downtime and previous gap value into the gap prediction model. In this case, the control device 200 can obtain the difference between the start time of the current roll forming process and the end time of the previous roll forming process as the downtime.

[0116] After performing operation S1012, the control device 200 outputs the predicted gap value, the previous gap value, and the downtime via the user interface 230 (operation S1014), and controls the servo valves of the second roller position adjustment units 130, 140, and 150, so that the roller gap between the first roller 110 and the second roller 120 becomes the predicted gap value. The control device 200 then adjusts the position of the second roller 120 (operation S1016). The control device 200 can control the servo valves to adjust the position of the second roller 120 based on the predicted gap value, thereby controlling the gap between the first roller 110 and the second roller 110.

[0117] After performing operation S1016, the control device 200 uses the thickness sensor 180 to measure the thickness of the first electrode after rolling (operation S1018) and determines whether the measured thickness of the first electrode is within the preset reference thickness range (operation S1020).

[0118] When the thickness of the first electrode is within the reference thickness range as determined in operation S1020, the control device 200 derives the thickness trend of the electrode 10 (operation S1022). Figure 11 As shown in the figure, when the gap prediction model is used to adjust the roll gap, it can be seen that the thickness of electrode 10 is more uniform compared with manually adjusting the roll gap.

[0119] When the thickness of the first electrode deviates from the reference thickness range as determined in operation S1020, the control device 200 adjusts the roller gap (operation S1024) and executes operation S1018.

[0120] Figure 12 This is a flowchart of a method for generating a gap prediction model according to an embodiment of the present disclosure.

[0121] refer to Figure 12 Server 300 collects process-related data, including standard data, exported data, and process data for a past set time (operation S1102).

[0122] When operation S1102 is executed, server 300 performs preprocessing on process-related data (operation S1104). Here, preprocessing may include data merging, repeating time processing, data sorting, missing value handling, outlier handling, etc.

[0123] When operation S1104 is executed, server 300 selects the factors necessary for modeling by analyzing the correlation between preprocessed process-related data and electrode thickness (operation S1106).

[0124] Processor 340 can select the necessary factors for modeling by analyzing the process influence and the correlation between various factors. That is, by using electrode thickness as the output characteristic and the remaining data after excluding electrode thickness from the process-related data as the input characteristic, processor 340 can select the necessary factors for modeling by analyzing the correlation between the input and output characteristics. In this case, processor 340 can use feature importance techniques to analyze the correlation between the input and output characteristics. For example, by using the y-factor as electrode thickness and the x-factor as the remaining data after excluding electrode thickness from the process-related data, processor 340 can select a certain number of factors in order of their greatest impact on electrode thickness.

[0125] After performing operation S1106, server 300 uses the selected factors as input and the roll gap value as output to generate a gap prediction model (operation S1108). In this case, server 300 can use a regression equation based on an exponential function to generate the gap prediction model, which uses the factors selected from process-related data as input and outputs the roll gap value.

[0126] According to this disclosure, a gap prediction model capable of predicting the roll gap value is generated, and the generated gap prediction model is used to predict the optimal roll gap value. This allows for minimizing quality deviations in the electrode rolling process.

[0127] The embodiments described herein can be implemented as, for example, methods or processes, apparatus, software programs, data streams, or signals. Although discussed in the context of a single type of implementation (e.g., discussed only as a method), the features discussed herein can also be implemented in other forms (e.g., apparatus or program). The apparatus can be implemented by suitable hardware, software, firmware, etc. The method can be implemented on an apparatus such as a processor, which generally refers to a processing apparatus including computers, microprocessors, integrated circuits, programmable logic devices, etc. The processor can include communication devices such as computers, cellular phones, personal digital assistants (PDAs), and other means that facilitate information communication between the device and the end user.

[0128] Although this disclosure has been described with reference to embodiments and accompanying drawings illustrating various aspects thereof, this disclosure is not limited thereto. Various modifications and variations can be made by those skilled in the art within the scope of the technical spirit of this disclosure.

Claims

1. A rolling mill for manufacturing electrodes, the rolling mill comprising: A roll forming unit includes a first roll and a second roll, the first roll and the second roll being configured to form a roll forming gap, through which the electrode passes between the first roll and the second roll; The roller position adjustment unit is configured to adjust the position of the second roller; as well as The control device is configured to use a gap prediction model to predict the roll gap value and control the roll position adjustment unit to make the roll gap change to the predicted roll gap value.

2. The roller pressing equipment according to claim 1, wherein, The roller position adjustment unit includes: A drive unit, connected to the second roller; and A control valve is configured to receive the predicted roll gap value from the control device and control the drive unit to adjust the position of the second roller such that the roll gap becomes the predicted roll gap value.

3. The roller pressing equipment according to claim 1, further comprising a user interface, in, When a conditional inference command is input through the user interface, the control device inputs the downtime and previous gap value into the gap prediction model to predict the roll gap value.

4. The roller pressing equipment according to claim 3, wherein, The downtime represents the difference between the start time of the current roll pressing process and the end time of the previous roll pressing process.

5. The roller pressing equipment according to claim 1, further comprising: A thickness sensor is configured to measure the thickness of the electrode passing between the first roller and the second roller.

6. The roller pressing equipment according to claim 5, wherein, The control device compares the thickness of the electrode measured by the thickness sensor with a preset reference thickness range, and adjusts the roller gap when the thickness of the electrode exceeds the reference thickness range.

7. The roller pressing equipment according to claim 1, further comprising: A server is configured to collect process-related data, including at least one of standard data, derived data, and process data for a specified past time period, and the server is configured to use the collected process-related data to generate the gap prediction model.

8. The roller pressing equipment according to claim 7, wherein, By using the process-related data as input characteristics and the electrode thickness as output characteristics, the server selects the factors necessary for modeling by analyzing the correlation between the input characteristics and the output characteristics, applies the selected factors to each of a plurality of machine learning models to predict the roll gap value, evaluates the performance of each machine learning model based on the predicted roll gap value, selects one of the machine learning models based on the performance of each machine learning model, and provides the selected machine learning model as the gap prediction model.

9. The roller pressing equipment according to claim 8, wherein, The server uses feature importance techniques to analyze the correlation between the input characteristics and the output characteristics.

10. The roller pressing equipment according to claim 8, wherein, The server uses a regression equation based on an exponential function to generate the gap prediction model, which uses the selected factors as inputs and outputs the roll gap value.

11. A method of operating a rolling mill for manufacturing an electrode, the rolling mill comprising a first roller and a second roller forming a rolling gap through which the electrode passes, and a control device for controlling the adjustment of the position of the second roller, the method comprising: Use a gap prediction model to predict the roll gap value; as well as The roller position adjustment unit is controlled to adjust the position of the second roller so that the roller gap becomes the predicted roller gap value.

12. The method according to claim 11, wherein, In the prediction of the roll gap value, when a condition inference command is input through the user interface of the roll forming equipment, the control device inputs the downtime and previous gap value into the gap prediction model to predict the roll gap value.

13. The method according to claim 12, wherein, The downtime represents the difference between the start time of the current roll pressing process and the end time of the previous roll pressing process.

14. The method of claim 11, further comprising: After the adjustment of the position of the second roller, the thickness of the electrode pressed by the first roller and the second roller is compared with a preset reference thickness range, and the roller gap is adjusted when the thickness of the electrode deviates from the reference thickness range.

15. The method of claim 11, further comprising: Prior to the prediction of the roll gap value, process-related data including at least one of standard data, derived data, and process data for a past set time period are collected; as well as The gap prediction model is generated using the collected process-related data.

16. The method according to claim 15, wherein, In the generation of the gap prediction model that uses the process-related data as input characteristics and electrode thickness as output characteristics, the necessary factors for modeling are selected by analyzing the correlation between the input characteristics and the output characteristics. The selected factors are applied to each of a plurality of machine learning models to predict the roll gap value. The performance of each machine learning model is evaluated based on the predicted roll gap value. One of the machine learning models is selected based on the performance of each machine learning model, and the selected machine learning model is generated as the gap prediction model.

17. The method according to claim 16, wherein, In the generation of the gap prediction model, feature importance techniques are used to analyze the correlation between the input characteristics and the output characteristics.

18. The method according to claim 15, wherein, In the generation of the gap prediction model, a regression equation based on an exponential function is used to generate the gap prediction model, the regression equation using the selected factors as input and outputting the roll gap value.

19. The method of claim 11, further comprising: Before predicting the roll gap value, the gap prediction model is received from the server and stored.

20. The method according to claim 19, wherein, The server collects process-related data, including at least one of standard data, exported data, and process data for a past set time, and uses the collected process-related data to generate the gap prediction model.

Images