Film sticking device and method for flexible curved film

Abstract

The present application relates to the technical field of flexible curved film application, and discloses a film application device for a flexible curved film, comprising a first application component, a second application component, a controller, and a drive component. The first application component comprises a first silicone pad, a first support, and an adsorption component, wherein the first silicone pad and the adsorption component are fixedly connected to the bottom of the first support, and a plurality of first flexible tubes are disposed within the first silicone pad. The second application component comprises a second silicone pad and a second support, wherein the second silicone pad is fixedly connected to the top of the second support, and a plurality of second flexible tubes are disposed within the second silicone pad. The first and second silicone pads are disposed relative to each other, and the drive component drives the first application component to move closer to or further from the second application component. Each first and second flexible tube is connected to a working pump via an electrically controlled valve. A working fluid is passed through the first and second flexible tubes. This application can improve the accuracy and effectiveness of film application.

Description

Technical Field

[0001] The present application relates to the technical field of flexible curved surface film lamination, and more specifically, to a film lamination device and method for flexible curved surface films. Background Art

[0002] As 3C products are frequently used in daily life, their comfort has become the focus of major manufacturers in expanding product demand. Due to their higher image clarity, flexible organic electroluminescence display (OLED) screens are gradually replacing liquid crystal display (LCD) screens in traditional smart terminal products. In order to increase the screen-to-body ratio of OLED screens in terminal products, the screen cover glass of some terminal products is made into a 3D shape, thereby creating a more comfortable and wide visual experience for users. This ergonomic design greatly improves the user experience. For example, the screen of a terminal product made of 3D curved glass has the advantages of being light, clean, anti-glare and highly weather-resistant, which also greatly improves the screen-to-body ratio of the terminal product. In addition to the huge market prospects in terminal products such as smartphones, 3D curved glass screens can also be used in in-vehicle smart terminals, head-mounted VR devices and portable instrument panels, and will also be used in wearable smart mobile terminal products such as smart bracelets.

[0003] The lamination process for curved screens involves tightly applying an optical film to the glass surface. Generally speaking, the curvature of curved screens places higher demands on the lamination technology, especially for small and medium-sized curved screens. More importantly, during the lamination process, it is necessary to ensure a tight fit between the flexible film and the more brittle screen, while also ensuring that excessive force is not applied to avoid damaging the screen. Therefore, there is an urgent need for an efficient and high-precision device for laminating flexible curved surfaces to improve both lamination quality and accuracy. Summary of the Invention

[0004] The purpose of this application is to provide a film-sticking device and method for flexible curved films, which solves the technical problems of low film-sticking efficiency and low precision of existing flexible curved films, and achieves the technical effect of improving the film-sticking efficiency and precision of flexible curved films.

[0005] In the first aspect, the embodiment of the present application provides a film-sticking device for a flexible curved film, including a first bonding component, a second bonding component, a controller and a drive component, the first bonding component including a first silicone pad, a first support and an adsorption component, the first silicone pad and the adsorption component are fixedly connected to the bottom of the first support, a plurality of first flexible tubes are provided in the first silicone pad, and the adsorption component is used to adsorb the flexible curved film; the second bonding component includes a second silicone pad and a second support, the second silicone pad is fixedly connected to the top of the second support, a plurality of second flexible tubes are provided in the second silicone pad; the first silicone pad and the second silicone The pads are arranged relative to each other, and the driving component drives the first fitting component and the second fitting component to be relatively close or away. Each first flexible tube and each second flexible tube is connected to a working pump through an electric control valve, and the working pump and each electric control valve are electrically connected to the controller respectively; the first flexible tube and the second flexible tube are used to pass the working fluid, and a plurality of displacement sensors are respectively provided in the first silicone pad and / or the second silicone pad. The controller controls the opening and closing state of each electric control valve and the power of the working pump in real time according to the position value collected by the position sensor, so as to control the extrusion state of the first silicone pad and / or the second silicone pad on the adsorption flexible curved membrane.

[0006] In a possible implementation of the first aspect, the controller includes an LSTM prediction module, which determines the predicted position corresponding to the target point at the predicted moment based on the current moment value, the current position of the target point on the first silicone pad and / or the second silicone pad, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube corresponding to the target point, the pressure value in the second flexible tube corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad, and the type of the second silicone pad, and determines the control quantity corresponding to the target point at the predicted moment based on the predicted position, and controls the opening and closing state of the electric control valve corresponding to the target point and the real-time power of the working pump based on the control quantity.

[0007] In another possible implementation of the first aspect, the controller also includes a fuzzy control module, which determines a target position compensation value of the predicted position and the theoretical position based on the predicted position and the theoretical position corresponding to the target point at the predicted moment. The fuzzy control module determines a compensation power value based on the target position compensation value, and adjusts the real-time power of the working pump based on the power compensation value.

[0008] In another possible implementation of the first aspect, the theoretical position corresponding to the target point at the prediction moment is determined as follows: the position change curve of the target point is determined through simulation experiments and / or fitting experiments; the position change curve is divided according to a preset interval value to determine the position curve segment corresponding to each time period; based on the position curve segment, the predicted value of the target point in each time period under discrete control is determined; and the predicted value corresponding to the position curve segment corresponding to the prediction moment is used as the theoretical position.

[0009] In another possible implementation of the first aspect, the controller controls the opening and closing state of each electrically controlled valve and the power of the working pump in real time through a PID algorithm.

[0010] In another possible implementation of the first aspect, the controller further includes a temperature response module, and a plurality of temperature sensors are respectively provided in the first silicone pad and / or the second silicone pad. When the first bonding component and the second bonding component are relatively far apart, the temperature response module controls the opening and closing state of each electrically controlled valve in real time based on the temperature value collected by the temperature sensor and the temperature value of the working fluid in each first flexible tube and each second flexible tube to adjust the temperature value of the first silicone pad and / or the second silicone pad.

[0011] In another possible implementation of the first aspect, the temperature response module is a neural network prediction module, which determines the temperature difference between the target temperature value after temperature control and the predicted temperature value based on the temperature value collected by the temperature sensor and the temperature difference between the target temperature value, the actual temperature value of the current working fluid, the flow value of the current working fluid, the pressure value of the current working fluid and the temperature control duration value.

[0012] In another possible implementation manner of the first aspect, the temperature response module and the controller perform time-division multiplexing control on each first flexible pipe and each second flexible pipe.

[0013] In a second aspect, an embodiment of the present application also provides a method for attaching a flexible curved film, the method comprising: adsorbing and connecting the flexible curved film on a first silicone pad, placing a screen on a second silicone pad, and driving the flexible curved film to attach to the screen on the second silicone pad through the first silicone pad, wherein a plurality of first flexible tubes are provided in the first silicone pad, and a plurality of second flexible tubes are provided in the second silicone pad, and a working fluid is simultaneously introduced into the first flexible tube and the second flexible tube; determining a predicted position corresponding to the target point at a predicted time according to a current time value, a current position of a target point on the first silicone pad and / or the second silicone pad, a position value of a first neighboring point of the target point, a position value of a second neighboring point of the target point, a position value of a third neighboring point of the target point, a pressure value in the first flexible tube corresponding to the target point, a pressure value in the second flexible tube corresponding to the target point, an external temperature value, a duration of the bonding process, a type of the first silicone pad, and a type of the second silicone pad; determining a control quantity corresponding to the target point at the predicted time according to the predicted position; and controlling the opening and closing state of an electrically controlled valve corresponding to the target point and the real-time power of a working pump according to the control quantity.

[0014] In a possible implementation of the second aspect, the method further includes: determining a target position compensation value based on the predicted position and the theoretical position corresponding to the target point at the predicted moment; determining a compensation power value of the working pump based on the target position compensation value; and adjusting the real-time power of the working pump based on the power compensation value.

[0015] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0016] The invention comprises a first bonding component and a second bonding component, the first bonding component comprises a first silicone pad, the second bonding component comprises a second silicone pad, the first silicone pad and the second silicone pad respectively fix the flexible protective film and the screen glass, the first silicone pad is provided with a first flexible tube capable of controlling the shape and pressure of each position of the first silicone pad, the second silicone pad is provided with a second flexible tube capable of controlling the shape and pressure of each position of the second silicone pad, by controlling the shape and pressure of each position of the first silicone pad and the second silicone pad, the shape and pressure at each position of the flexible protective film and the screen glass can be controlled, thereby improving the precision between the flexible protective film and the screen glass, and improving the attachment efficiency of the flexible protective film and the screen glass. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 1 is a schematic diagram of the three-dimensional structure of a film-applying device for a flexible curved film provided in an embodiment of the present application;

[0019] Figure 2 1 is a schematic diagram of the top view of the second silicone pad in the embodiment of the present application;

[0020] Figure 3 yes Figure 2 A schematic structural diagram of the AA section of the second silicone pad;

[0021] Figure 4 This is a schematic diagram of a control structure of a film-applying device for a flexible curved film provided in an embodiment of the present application;

[0022] Figure 5 is a schematic structural diagram of a controller in an embodiment of the present application;

[0023] Figure 6 1 is a schematic diagram of the top view of the second silicone pad in the embodiment of the present application;

[0024] Figure 7 Schematic diagram of the process of determining the theoretical position in the embodiment of the present application;

[0025] Figure 8 2 is a schematic diagram showing the distribution of the position measurement points of the second silicone pad in the embodiment of the present application;

[0026] Figure 9 1 is a schematic diagram of a module of a film-applying device for a flexible curved film in an embodiment of the present application;

[0027] Figure 10 Schematic diagram of a sample for training an LSTM prediction module in an embodiment of the present application;

[0028] Figure 11 Schematic diagram of the structure of the training temperature response module in an embodiment of the present application;

[0029] In the figure, 100, the first bonding component; 110, the first silicone pad; 111, the first flexible tube; 120, the first support; 130, the adsorption component; 200, the second bonding component; 210, the second silicone pad; 211, the second flexible tube; 220, the second support; 300, the controller; 301, the LSTM prediction module; 302, the fuzzy control module; 303, the temperature response module; 310, the electric control valve; 320, the displacement sensor; 330, the temperature sensor; 400, the drive component; 500, the working pump. DETAILED DESCRIPTION

[0030] In order to make the technical problems, technical solutions and beneficial effects to be solved by this application more clearly understood, this application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0031] It should be noted that when a component or structure is referred to as being “fixed to” or “disposed on” another component or structure, it may be directly on the other component or structure or indirectly on the other component or structure. When a component or structure is referred to as being “connected to” another component or structure, it may be directly connected to the other component or structure or indirectly connected to the other component or structure.

[0032] It should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the method or a component or structure referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0033] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0034] The lamination process for curved screens involves tightly applying an optical film to the glass surface. Generally speaking, the curvature of curved screens places higher demands on the lamination technology, especially for small and medium-sized curved screens. More importantly, the lamination process requires a tight fit between the flexible film and the more brittle screen, while also ensuring that excessive force is applied to avoid damaging the screen.

[0035] Based on the above reasons, an embodiment of the present application provides a film-laminating device for a flexible curved film, comprising a first laminating component and a second laminating component, the first laminating component comprising a first silicone pad, the second laminating component comprising a second silicone pad, the first silicone pad and the second silicone pad respectively fixing the flexible protective film and the screen glass, the first silicone pad being provided with a first flexible tube capable of controlling the shape and pressure of various positions of the first silicone pad, the second silicone pad being provided with a second flexible tube capable of controlling the shape and pressure of various positions of the second silicone pad, by controlling the shape and pressure of various positions of the first silicone pad and the second silicone pad, the shape and pressure at various positions of the flexible protective film and the screen glass can be controlled, thereby improving the precision between the flexible protective film and the screen glass, and improving the attachment efficiency of the flexible protective film and the screen glass.

[0036] The following describes a film-sticking device for a flexible curved film provided in an embodiment of the present application with reference to specific examples.

[0037] The present application provides a device for attaching a flexible curved film, comprising a first attaching assembly 100, a second attaching assembly 200, a controller 300, and a drive assembly 400. The first attaching assembly 100 comprises a first silicone pad 110, a first support 120, and an adsorption assembly 130. The first silicone pad 110 and the adsorption assembly 130 are fixedly connected to the bottom of the first support 120. The first silicone pad 110 is provided with a plurality of first flexible tubes 111. The adsorption assembly 130 is used to adsorb the flexible curved film. The second attaching assembly 200 comprises a second silicone pad 210 and a second support 220. The second silicone pad 210 is fixedly connected to the top of the second support 220. The second silicone pad 210 is provided with a plurality of second flexible tubes 211. The first silicone pad 110 and the second silicone pad 210 are arranged relative to each other. The driving assembly 400 drives the first bonding assembly 100 and the second bonding assembly 200 to move closer or farther away from each other. Each first flexible tube 111 and each second flexible tube 211 is connected to a working pump 500 via an electrically controlled valve 310. The working pump 500 and each electrically controlled valve 310 are electrically connected to the controller 300. The first flexible tube 111 and the second flexible tube 211 are used to pass a working fluid. Multiple displacement sensors 320 are respectively provided in the first silicone pad 110 and / or the second silicone pad 210. The controller 300 controls the opening and closing state of each electrically controlled valve 310 and the power of the working pump 500 in real time based on the position values ​​collected by the position sensors 320, thereby controlling the squeezing state of the first silicone pad 110 and / or the second silicone pad 210 on the adsorbent flexible curved membrane.

[0038] Specifically, Figure 1 : is a schematic diagram of a three-dimensional structure of a film-applying device for a flexible curved film provided in an embodiment of the present application, such as Figure 1As shown, the first bonding component 100 is used to position the flexible curved film, and the first bonding component 100 can attach the flexible curved film to the screen glass when it moves; the second bonding component 200 is used to position the screen glass, and the first bonding component 100 and the second bonding component 200 can cooperate with each other to attach the flexible curved film to the screen glass.

[0039] Specifically, the controller 300 is used to centrally control the film laminating device for a flexible curved film, and the drive assembly 400 is used to drive the first laminating assembly 100 and the second laminating assembly 200 to move closer to or farther from each other. The drive assembly 400 can be a cylinder, and the first laminating assembly 100 can be slidably connected to a guide rail. The first laminating assembly 100 can be driven by the drive assembly 400 to move closer to or farther from the second laminating assembly 200.

[0040] Specifically, the adsorption component 130 is used to adsorb the flexible curved membrane. The adsorption component 130 is an array microporous structure. The adsorption component 130 of the array structure is connected to the vacuum chamber through a pipeline, so that the adsorption component 130 can fix the flexible curved membrane.

[0041] Specifically, when the working medium is filled into the first flexible tube 111, the first flexible tube 111 can expand inside the first silicone pad 110 to adjust the shape and pressure of the first silicone pad 110. When the working medium is filled into the second flexible tube 211, the second flexible tube 211 can expand inside the second silicone pad 210 to adjust the shape and pressure of the second silicone pad 210.

[0042] Specifically, for small and medium-sized 3C curved surface components, for example, when applying film to 4 to 6-inch smartphone glass cover plates, the number of first flexible tubes 111 can be set to 12 to 18, and the number of second flexible tubes 211 can be set to 16 to 24.

[0043] Specifically, a plurality of displacement sensors 320 are respectively provided in the first silicone pad 110 and / or the second silicone pad 210. The controller 300 controls the opening and closing state of each electric control valve 310 and the power of the working pump 500 in real time according to the position value collected by the position sensor 320, so as to control the extrusion state of the first silicone pad 110 and / or the second silicone pad 210 on the adsorption flexible curved membrane.

[0044] For example, the plurality of displacement sensors 320 may be disposed only in the second silicone pad 210 , and the plurality of displacement sensors 320 may be capable of detecting the deformation amount at various positions of the second silicone pad 210 .

[0045] It should be noted that the multiple displacement sensors 320 in the embodiment of the present application can also be set in the first silicone pad 110, or can be set in the first silicone pad 110 and the second silicone pad 210 at the same time. The embodiment of the present application does not limit the setting position of the displacement sensors 320 in the first silicone pad 110 and the second silicone pad 210.

[0046] Figure 2 : is a schematic diagram of the top view of the second silicone pad in the embodiment of the present application, Figure 3 yes Figure 2 A schematic structural diagram of the AA section of a second silicone pad, as shown in FIG. Figure 2 and Figure 3 As shown, the plurality of displacement sensors 320 can detect the displacement of each position of the second silicone pad 210. At the same time, when the working medium is introduced into the second flexible tube 211, the shape of the second silicone pad 210 can be adjusted.

[0047] For example, the working medium may be a common gas or liquid.

[0048] Figure 4 Schematic diagram of the control structure of a film-attaching device for a flexible curved film provided in an embodiment of the present application. Figures 1 to 4 As shown, when the film-laminating device for the flexible curved film in the embodiment of the present application is in use, the driving component 400 drives the first laminating component 100 to approach the second laminating component 200, the first silicone pad 110 fixes the flexible curved film, and the second silicone pad 210 is used to fix the screen glass. When laminating the flexible curved film, the flexible curved film starts from the center line of the first silicone pad 110 and gradually adheres to the screen glass on the second silicone pad 210 from the center line of the first silicone pad 110 to the two sides of the first silicone pad 110, thereby realizing the attachment of the flexible curved film.

[0049] When the flexible curved film is attached in the embodiment of the present application, the deformation amount of each position of the flexible curved film can be adjusted by the second flexible tube 211 in the second silicone pad 210. When the second flexible tube 211 increases the deformation amount, the second flexible tube 211 can increase the squeezing amount of the second silicone pad 210 on the flexible curved film, so as to increase the fitting strength between the flexible curved film and the screen glass at this position. The embodiment of the present application improves the fitting accuracy and efficiency of the flexible curved film by precisely controlling the strength and squeezing amount of the flexible curved film at each position.

[0050] The embodiment of the present application has multiple independent force and position combined control modules built into the first silicone pad and / or the second silicone pad, which can independently control different areas when the flexible curved film is being laminated, and laminate according to the flexible curved film lamination process requirements, thereby improving the lamination accuracy of the flexible curved film.

[0051] In some implementations, the controller 300 includes an LSTM prediction module 301, which determines the predicted position corresponding to the target point at the predicted moment based on the current moment value, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad 110 and the type of the second silicone pad 210, and determines the control quantity corresponding to the target point at the predicted moment based on the predicted position, and controls the opening and closing state of the electric control valve 310 corresponding to the target point and the real-time power of the working pump 500 based on the control quantity.

[0052] When laminating the flexible protective film, in order to improve the effect of discharging bubbles when the flexible protective film is laminating to the screen glass, the embodiment of the present application gradually controls the displacement of each position on the first silicone pad 110 and / or the second silicone pad 210 through the LSTM prediction module 301, so that the target point on the first silicone pad 110 and / or the second silicone pad 210 can be laminating to the screen glass under the control of the LSTM prediction module 301.

[0053] In the embodiment of the present application, the LSTM prediction module 301 predicts the predicted position corresponding to the target point at the predicted moment based on the current moment value, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad 110 and the type of the second silicone pad 210, and then determines the control amount corresponding to the target point at the predicted moment according to the predicted position, and controls the opening and closing state of the electric control valve 310 corresponding to the target point and the real-time power of the working pump 500 according to the control amount, so that the embodiment of the present application can accurately control the position of the target point and improve the bonding accuracy of the flexible curved membrane.

[0054] Specifically, the LSTM prediction module 301 predicts the predicted position corresponding to the target point at the predicted time based on the current moment value of the target point, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad 110 and the type of the second silicone pad 210 through the LSTM algorithm.

[0055] When fitting the flexible curved membrane, due to the complex mold space structure of the flexible curved membrane, the embodiment of the present application improves the problem of low position accuracy of the second silicone pad collected by a single position sensor. The deformation variables at different positions on the second silicone pad are collected by multiple pressure sensors, and then the position data of multiple adjacent position sensors collected in space are used to predict the position of the flexible curved membrane at a future moment, thereby improving the fitting accuracy of the flexible curved membrane.

[0056] In some implementations, the controller 300 also includes a fuzzy control module 302, which determines a target position compensation value of the predicted position and the theoretical position based on the predicted position and the theoretical position corresponding to the target point at the predicted moment. The fuzzy control module 302 determines a compensation power value based on the target position compensation value, and adjusts the real-time power of the working pump 500 based on the power compensation value.

[0057] The embodiment of the present application determines the target position compensation value based on the predicted position and the theoretical position corresponding to the target point at the predicted moment, and then adjusts the real-time power of the working pump 500, thereby achieving the control effect of improving the position of the target point and improving the control accuracy when the flexible curved membrane is attached. In order to solve the problem of low position control accuracy when the flexible protective film is attached, the embodiment of the present application can integrate the position deviation at the current moment and the position deviation at the future moment, control the pressure and position of the film, and improve the attachment accuracy of the flexible curved membrane. At the same time, the embodiment of the present application performs real-time online control of the film attachment process through the fuzzy control module 302, which not only improves the control accuracy of the system, but also improves the dynamic performance of the system position control process.

[0058] Specifically, when determining the target position compensation value of the predicted position and the theoretical position, the difference between the predicted position and the theoretical position may be used as the target position compensation value.

[0059] Figure 5 is a schematic diagram of the structure of the controller in the embodiment of the present application, such as Figure 5As shown, the embodiment of the present application combines the LSTM prediction module 301 and the fuzzy control module 302, and can jointly control the fitting process of the flexible curved film through the LSTM prediction module 301 and the fuzzy control module 302, thereby improving the fitting accuracy of the flexible curved film and improving the fitting effect of the flexible curved film.

[0060] In some implementations, Figure 7 Schematic diagram of the process of determining the theoretical position in the embodiment of the present application. Figure 7 As shown, determining the predicted position corresponding to the target point at the predicted moment includes S710 to S740, and S710 to S740 are described in detail below.

[0061] S710: Determine a position change curve of a target point through simulation experiments and / or fitting experiments.

[0062] In the embodiment of the present application, after a simulation experiment and / or a fitting experiment, the optimal fitting position change curve of the first silicone pad or the second silicone pad can be determined.

[0063] S720: Segment the position change curve according to a preset interval value to determine a position curve segment corresponding to each time period.

[0064] The preset interval value may be 0.1 to 0.3 seconds. After obtaining the position change curve, the position change curve is segmented into intervals of 0.1 to 0.3 seconds.

[0065] S730. Determine the predicted value of the target point in each time period under discrete control according to the position curve segment.

[0066] In the embodiment of the present application, the predicted value corresponding to the target point under each position curve segment can be obtained through the position curve segment and the predicted value of the target point in each time period under discrete control.

[0067] S740: Use the predicted value corresponding to the position curve segment corresponding to the predicted time as the theoretical position.

[0068] Through simulation experiments, the embodiment of the present application can successfully determine the theoretical position corresponding to the target point at the prediction moment.

[0069] Figure 9 Schematic diagram of a module of a film-attaching device for a flexible curved film in an embodiment of the present application. Figure 9 As shown, the collaborative control process of the film-attaching device for the flexible curved film in the embodiment of the present application is specifically described below with reference to a specific example, which specifically includes the following steps:

[0070] (1) In the theoretical model module, the ideal change curves of each position measurement point of the second silicone pad are segmented at intervals of 0.1 - 0.3 seconds, and segmented discrete control is implemented for each time region in sequence. For the position control of the [tm, tn] time region of the i-th position measurement point, the set position is TSi(tn); in the instantaneous inspection module, the instantaneous position of this position measurement point after the second flexible tube corresponding to the time region [tm, tn] is pressurized is detected and denoted as ISi(t).

[0071] (2) The LSTM prediction module collects multiple groups of feature vectors (the feature vectors specifically include the current moment value, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position values of the first neighboring point, the second neighboring point, and the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the fitting process, the type of the first silicone pad 110, and the type of the second silicone pad 210).

[0072] Figure 8 It is a schematic diagram of the distribution of the position measurement points of the second silicone pad in the embodiment of the present application. As Figure 8 shown, for the PD1 position measurement point of the second silicone pad, its current adjacent point 1, current adjacent point 2, and current adjacent point 3 are respectively the PD2 position measurement point, PD3 position measurement point, and PD4 position measurement point of the second silicone pad; for the PD2 position measurement point of the second silicone pad, its current adjacent point 1, adjacent point 2, and adjacent point 3 are respectively the PD1 position measurement point, PD3 position measurement point, and PD4 position measurement point of the second silicone pad.

[0073] (3) The instantaneous feature vectors collected in the [tm, tn] time region of the i-th position measurement point are input into the trained LSTM prediction module to obtain the position output of this measurement point at the future tn moment, denoted as PSi(tn).

[0074] (4) The fuzzy control module takes the LSTM predicted position PSi(tn) of the i-th position measurement point at the future tn moment as the input, and uses the fuzzy control method to perform a forward-looking adjustment on the position of this position measurement point, and the output is the regulation amount YSi(t) at the current moment.

[0075] (5) Calculate the instantaneous position difference eSi(t) through formula (1),

[0076] eSi(t) = TSi(tn) - ISi(t) (1)

[0077] In formula (1), after eSi(t) is input into the controller, the current control quantity XSi(t) corresponding to the i-th position measurement point in the controller is obtained. The control mode of the controller 300 can be a PID control mode.

[0078] (6) The controller combines the current control quantity XSi(t) with the current control quantity YSi(t) of the fuzzy control module to control the position and pressure value of the second flexible pipe.

[0079] It should be understood that the position collaborative control method for other position measurement points is similar to the above steps (1) to (6) and will not be repeated here.

[0080] like Figure 9 As shown, through the cooperation of the LSTM prediction module, the fuzzy control module and the controller, the above steps (1) to (6) can be completed, thereby achieving the effect of improving the fitting accuracy and fitting efficiency of the flexible curved film.

[0081] The embodiment of the present application integrates PID control and fuzzy control to achieve precise control of the film sticking force and film sticking position in the specific area to which it is applied (spatial coordination), thereby improving the film sticking accuracy and efficiency.

[0082] In some implementations, the controller 300 controls the opening and closing state of each electrically controlled valve 310 and the power of the working pump 500 in real time through a PID algorithm.

[0083] Specifically, the PID algorithm of the controller 300 includes a proportional unit (P), an integral unit (I) and a differential unit (D). The PID algorithm controls the input of the i-th position measurement point eS i (t) and output XS i (t) is expressed by formula (1):

[0084]

[0085] In formula (2), eU i(t) represents the error term, represents the error integral term, represents the error differential term, k iUp Indicates the proportional coefficient of the i-th position measurement point, k iUI Indicates the integral coefficient of the i-th position measurement point, k iUD Represents the differential coefficient of the i-th measurement point.

[0086] For example, in the fuzzy control module of the embodiment of the present application, the LSTM predicted position PS of the i-th position measurement point at the next tn moments is calculated as follows: i(tn) is used as input, and the fuzzy control method is used to adjust the position of the position measurement point in advance, and the output is the current moment control value YS i (t). The fuzzy control strategy of the fuzzy control module is represented by Table 1. As shown in Table 1, taking one of the examples as an example, when the expected position TS at the future time tn is i (tn)∈(0~0.1mm], and ABS(TS i (tn)-PS i (tn))≤0.01mm, if TS i (tn) is greater than PS i (tn), then the power adjustment rate of the working pump connected to the second flexible pipe controlling the current position increases by 1%, that is, 101% of the standard power (preferably 200W / pipe). If TS i (tn) is less than PS i (tn), then its power regulation rate is reduced by 1%, that is, 99% of the standard power.

[0087] Table 1

[0088]

[0089]

[0090] For example, the training of the LSTM prediction module includes two parts: pre-training and post-training. Figure 10 Schematic diagram of a sample for training an LSTM prediction module in an embodiment of the present application, such as Figure 10 As shown, the samples of the pre-training and post-training parts include duration, features and batches; the duration is based on a sampling period of 40-100ms, and the entire bonding time is discretized to obtain quantitative data on the time scale. The batch size is 64 batches, of which 32 batches are pre-trained and 32 batches are post-trained. The features specifically include the current time value, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad 110 and the type of the second silicone pad 210.

[0091] In some implementations, pre-training data for the LSTM prediction module comes from simulation or experimental data.

[0092] In some implementations, when the post-training data batches of the LSTM prediction module are less than 32 batches, the feature vector of the current moment is randomly extracted from the pre-training 32 batches of data and replaced to meet the total number of batches of 32 batches, so that the LSTM prediction module can be trained normally.

[0093] In some implementations, the post-training 32 batches of data of the LSTM prediction module are divided into three update regions, namely the best prediction region, the worst prediction region, and the most recent prediction region. The best prediction region is 6 batches of data, the worst prediction region is 6 batches of data, and the most recent prediction region is 4 batches of data.

[0094] In some implementations, after a round of fitting is completed, the LSTM prediction module evaluates the predicted position of each discrete time point, calculates its prediction error, and stores it; the best prediction area is the batch with the best prediction error at the current moment after all the fittings, the worst prediction area is the batch with the worst prediction error at the current moment after all the fittings, and the most recent prediction area is the most recent batch.

[0095] In some implementations, the LSTM prediction module is implemented by an industrial control computer and its associated data storage, which stores pre-trained and post-trained sample data and its prediction model.

[0096] In some implementations, the LSTM prediction module trains the prediction model in real time in the background. When the prediction model training is completed, the current prediction model is replaced in the next round of fitting.

[0097] In some implementations, the controller 300 also includes a temperature response module 303, and multiple temperature sensors 330 are respectively provided in the first silicone pad 110 and / or the second silicone pad 210. When the first bonding component 100 and the second bonding component 200 are relatively far apart, the temperature response module 303 controls the opening and closing state of each electric control valve 310 in real time based on the temperature value collected by the temperature sensor 330 and the temperature value of the working fluid in each first flexible tube 111 and each second flexible tube 211 to adjust the temperature value of the first silicone pad 110 and / or the second silicone pad 210.

[0098] Figure 6 : is a schematic diagram of the top view of the second silicone pad in the embodiment of the present application. Figure 6 As shown, the embodiment of the present application can further improve the temperature control effect when laminating the flexible curved film by controlling the laminating temperature, thereby improving the laminating effect of the flexible curved film on the screen.

[0099] In some implementations, the temperature response module 303 is a neural network prediction module, which determines the temperature difference between the target temperature value after temperature control and the predicted temperature value based on the temperature value collected by the temperature sensor 330 and the temperature difference between the target temperature value, the actual temperature value of the current working fluid, the flow value of the current working fluid, the pressure value of the current working fluid and the temperature control duration value.

[0100] Figure 11 Schematic diagram of the structure of the training temperature response module in the embodiment of the present application. Figure 11 As shown, the embodiment of the present application can improve the accuracy of temperature control and improve the effect of temperature control through the neural network prediction module.

[0101] Specifically, the training process of the temperature response module 303 includes:

[0102] First, a training set is constructed. The input parameters of the training set include the temperature difference ΔT between the temperature value T1 collected by the temperature sensor 330 and the target temperature value, the actual temperature value T2 of the current working fluid, the flow value Q1 of the current working fluid, the pressure value P1 of the current working fluid, and the temperature control duration value t. The output parameters of the training set include the temperature difference ΔT' between the target temperature value after temperature control and the predicted temperature value (the difference between the target temperature T1 and the current actual temperature).

[0103] Then, an input parameter level factor table is constructed; wherein, the factors in the input parameter level factor table include the temperature difference ΔT between the temperature value T1 and the target temperature value, the actual temperature value T2 of the current working fluid, the flow value Q1 of the current working fluid, the pressure value P1 of the current working fluid and the temperature control duration value t, and the input parameter level gear. The above factors can be set according to actual needs.

[0104] Exemplarily, pure water is used as the temperature control fluid for the experiment. Table 2 shows the input parameter level factor table for the neural network model training, as shown in Table 2.

[0105] Table 2

[0106]

[0107] Secondly, based on the current actual temperature T2 and target temperature T1 of the temperature-controlled fluid, the input parameters to be trained are selected in combination with the input parameter level factor table.

[0108] As shown in Table 2, the current temperature difference ΔT is divided into 7 levels: 1°C, 0°C, -1°C, -3°C, -5°C, -7°C and -10°C; the current fluid temperature T2 is divided into 5 levels: 10°C, 12°C, 15°C, 20°C and 25°C; the current fluid flow rate Q1 is divided into 5 levels: 50 ml / s, 100 ml / s, 200 ml / s, 400 ml / s and 800 ml / s; the current fluid pressure P1 is divided into 5 levels: 0.01 MPa, 0.05 MPa, 0.07 MPa, 0.11 MPa and 0.15 MPa; the temperature control duration t is divided into 3 levels: 1 s, 3 s and 5 s.

[0109] Based on the neural network, full-factor experimental data is pre-trained by inputting five different data. There are a total of 7x5x5x5x3=2625 groups, but not all rehearsal combinations require experimental analysis. For example: the current actual fluid temperature T2 is 15°C, and the difference temperature ΔT is -10°C. At this time and in this case, the adhesive bonding pad needs to be cooled quickly, so the current working fluid flow rate should be increased, that is, Q1 can be 400ml / s or 880ml / s, the current fluid pressure P1 can be 0.1Mpa or 0.15Mp, and the duration is 3s and 5s. Therefore, the actual number of experiments required is 2x2x2=8 groups.

[0110] Finally, the input parameters to be trained are predicted based on the neural network in turn, and the temperature difference with the smallest absolute value is selected as the optimal input parameter combination.

[0111] By analyzing the selected input parameters to be trained, the optimal temperature difference ΔT' after temperature control is obtained, where the closer the temperature difference ΔT' after temperature control is to 0, the better the effect; the processing based on the neural network is pre-trained according to the experiments selected in the full-factor experiment, and then by traversing Table 2, the temperature control parameters that conform to the actual situation are obtained, such as the current temperature difference ΔT, the current actual temperature of the temperature-controlled fluid T2, the current temperature-controlled fluid flow Q1, the current temperature-controlled fluid pressure P1 and the temperature control duration t, and real-time control is performed to achieve the expected temperature.

[0112] In some implementations, after the above steps, the method further includes:

[0113] Retraining is performed based on the predicted result data to improve the accuracy of the prediction; wherein the predicted result data is the prediction of the input parameters to be trained based on the neural network.

[0114] For example, the above-mentioned retraining based on the predicted result data to improve the prediction accuracy includes:

[0115] First, retain the input parameters and output parameters of the training set in the neural network.

[0116] Then, the input parameters and output parameters of the above neural network are evaluated. If the difference between the predicted temperature and the actual measured temperature is within ±1.5°C, it is marked as trainable data.

[0117] Secondly, the embodiment of the present application can group the trainable data according to the current differential temperature and the current temperature-controlled fluid temperature. Within each group, experimental data is evenly extracted based on the prediction accuracy, and retraining is performed based on the experimental data based on the neural network control. In this embodiment, a total of 120 groups of experimental data are used.

[0118] Finally, the embodiment of the present application can obtain a user-defined neural network for the user to choose. When the user turns on the machine, the user can choose the customized neural network to predict the input parameters of the temperature control.

[0119] In some implementations, the temperature response module 303 and the controller 300 perform time-division multiplexing control on each first flexible tube 111 and each second flexible tube 211 .

[0120] In the embodiment of the present application, the temperature response module 303 and the controller 300 perform time-division multiplexing control on each first flexible pipe 111 and each second flexible pipe 211 , thereby improving the control efficiency of each first flexible pipe 111 and each second flexible pipe 211 and improving the working efficiency of the temperature response module 303 and the controller 300 .

[0121] The embodiment of the present application further provides a method for applying a flexible curved film, the method comprising S1010 to S1040, and S1010 to S1040 are described in detail below.

[0122] S1010. Adsorb and connect the flexible curved membrane to the first silicone pad 110, place the screen on the second silicone pad 210, and drive the flexible curved membrane to adhere to the screen on the second silicone pad 210 through the first silicone pad 110. A plurality of first flexible tubes 111 are provided in the first silicone pad 110, and a plurality of second flexible tubes 211 are provided in the second silicone pad 210. At the same time, working fluid is introduced into the first flexible tubes 111 and the second flexible tubes 211.

[0123] By fixing the flexible curved film and the screen respectively, the first silicone pad 110 and the second silicone pad 210 can cooperate with each other to fit the flexible curved film and the screen.

[0124] S1020, determine the predicted position corresponding to the target point at the predicted time based on the current time value, the current position of the target point on the first silicone pad 110 and / or the second silicone pad 210, the position value of the first neighboring point of the target point, the position value of the second neighboring point of the target point, the position value of the third neighboring point of the target point, the pressure value in the first flexible tube 111 corresponding to the target point, the pressure value in the second flexible tube 211 corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad 110 and the type of the second silicone pad 210.

[0125] Specifically, after determining the predicted position corresponding to the target point at the predicted moment, the position of the flexible curved film and the screen can be accurately controlled, thereby improving the fitting accuracy of the flexible curved film and the screen.

[0126] S1030. Determine the control amount corresponding to the target point at the predicted time according to the predicted position.

[0127] Specifically, by determining the control amount at the prediction moment, the bonding process can be accurately controlled.

[0128] S1040 , controlling the opening and closing state of the electrically controlled valve 310 and the real-time power of the working pump 500 corresponding to the target point according to the control amount.

[0129] Through S1010 to S1040, the lamination accuracy of the flexible curved film and the screen can be improved, thereby improving the lamination efficiency of the flexible curved film and the screen.

[0130] In some implementations, the method further includes S1050 to S1070, which are described in detail below:

[0131] S1050: Determine a target position compensation value based on the predicted position and the theoretical position corresponding to the target point at the predicted moment.

[0132] Specifically, after the target position compensation value is determined, the pressure and position of the target point on the first silicone pad 110 and / or the second silicone pad 210 can be compensated and controlled according to the target position compensation value.

[0133] S1060 : Determine the compensation power value of the working pump 500 according to the target position compensation value.

[0134] By determining the compensation power value of the working pump 500 , the position of the target point of the first silicone pad 110 and / or the second silicone pad 210 can be compensated and controlled according to the pressure and position at the target point of the first silicone pad 110 and / or the second silicone pad 210 .

[0135] S1070 : Adjust the real-time power of the working pump 500 according to the power compensation value.

[0136] Through S1050 to S1070, the position of the target point on the first silicone pad 110 and / or the second silicone pad 210 can be accurately controlled, thereby improving the fitting accuracy and fitting efficiency of the flexible curved film and the screen.

[0137] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A film-sticking device for a flexible curved film, characterized in that: The invention comprises a first laminating component (100), a second laminating component (200), a controller (300) and a driving component (400); the first laminating component (100) comprises a first silicone pad (110), a first support (120) and an adsorption component (130); the first silicone pad (110) and the adsorption component (130) are fixedly connected to the bottom of the first support (120); a plurality of first flexible tubes (111) are provided in the first silicone pad (110); and the adsorption component (130) is used for adsorbing a flexible curved film; The second fitting component (200) comprises a second silicone pad (210) and a second support (220), the second silicone pad (210) being fixedly connected to the top of the second support (220), and a plurality of second flexible tubes (211) being provided in the second silicone pad (210); The first silicone pad (110) and the second silicone pad (210) are arranged relative to each other, and the driving component (400) drives the first fitting component (100) and the second fitting component (200) to move relatively closer or farther away. Each first flexible tube (111) and each second flexible tube (211) are connected to a working pump (500) via an electric control valve (310), and the working pump (500) and each electric control valve (310) are respectively electrically connected to the controller (300); The first flexible tube (111) and the second flexible tube (211) are used to pass a working fluid, and a plurality of displacement sensors (320) are respectively provided in the first silicone pad (110) and / or the second silicone pad (210). The controller (300) controls the opening and closing state of each electric control valve (310) and the power of the working pump (500) in real time according to the position value collected by the displacement sensor (320), so as to control the squeezing state of the first silicone pad (110) and / or the second silicone pad (210) on the adsorption flexible curved membrane; The controller (300) includes an LSTM prediction module (301), which determines a predicted position corresponding to the target point at a predicted moment based on a current moment value, a current position of a target point on the first silicone pad (110) and / or the second silicone pad (210), a position value of a first neighboring point of the target point, a position value of a second neighboring point of the target point, a position value of a third neighboring point of the target point, a pressure value in a first flexible tube (111) corresponding to the target point, a pressure value in a second flexible tube (211) corresponding to the target point, an external temperature value, a duration of a bonding process, a type of the first silicone pad (110), and a type of the second silicone pad (210), and determines a control amount corresponding to the target point at a predicted moment based on the predicted position, and controls the opening and closing state of the electric control valve (310) corresponding to the target point and the real-time power of the working pump (500) based on the control amount.

2. The film-sticking device for a flexible curved film according to claim 1, characterized in that: The controller (300) further includes a fuzzy control module (302), wherein the fuzzy control module (302) determines a target position compensation value between the predicted position and the theoretical position according to the predicted position and the theoretical position corresponding to the target point at the predicted moment, and the fuzzy control module (302) determines a compensation power value according to the target position compensation value, and adjusts the real-time power of the working pump (500) according to the power compensation value.

3. The film-sticking device for a flexible curved film according to claim 2, characterized in that: The theoretical position corresponding to the target point at the predicted moment is determined by: Determine the position change curve of the target point through simulation experiments and / or fitting experiments; Segmenting the position change curve according to a preset interval value to determine a position curve segment corresponding to each time period; determining a predicted value of the target point in each time period under discrete control according to the position curve segment; The predicted value corresponding to the position curve segment corresponding to the predicted moment is used as the theoretical position.

4. The film-sticking device for a flexible curved film according to claim 1, characterized in that: The controller (300) controls the opening and closing state of each electrically controlled valve (310) and the power of the working pump (500) in real time through a PID algorithm.

5. The film-sticking device for a flexible curved film according to claim 1, characterized in that: The controller (300) further includes a temperature response module (303), wherein a plurality of temperature sensors (330) are respectively provided in the first silicone pad (110) and / or the second silicone pad (210), and when the first fitting component (100) and the second fitting component (200) are relatively far apart, the temperature response module (303) controls the opening and closing state of each electric control valve (310) in real time based on the temperature value collected by the temperature sensor (330) and the temperature value of the working fluid in each first flexible tube (111) and each second flexible tube (211), so as to adjust the temperature value of the first silicone pad (110) and / or the second silicone pad (210).

6. The film-sticking device for a flexible curved film according to claim 5, characterized in that: The temperature response module (303) is a neural network prediction module. The temperature response module (303) determines the temperature difference between the target temperature value after temperature control and the predicted temperature value based on the temperature value collected by the temperature sensor (330) and the temperature difference between the target temperature value, the actual temperature value of the current working fluid, the flow value of the current working fluid, the pressure value of the current working fluid and the temperature control duration value.

7. The film-sticking device for a flexible curved film according to claim 5, characterized in that: The temperature response module (303) and the controller (300) perform time-division multiplexing control on each first flexible tube (111) and each second flexible tube (211).

8. A method for applying a flexible curved film, characterized in that: Using the film-sticking device for a flexible curved film according to any one of claims 1 to 7, the method comprises: The flexible curved film is adsorbed and connected to the first silicone pad (110), the screen is placed on the second silicone pad (210), and the flexible curved film is driven to adhere to the screen on the second silicone pad (210) through the first silicone pad (110), a plurality of first flexible tubes (111) are provided in the first silicone pad (110), and a plurality of second flexible tubes (211) are provided in the second silicone pad (210), and a working fluid is simultaneously introduced into the first flexible tubes (111) and the second flexible tubes (211); Determine the predicted position corresponding to the target point at the predicted time according to the current time value, the current position of the target point on the first silicone pad (110) and / or the second silicone pad (210), the position value of the first adjacent point of the target point, the position value of the second adjacent point of the target point, the position value of the third adjacent point of the target point, the pressure value in the first flexible tube (111) corresponding to the target point, the pressure value in the second flexible tube (211) corresponding to the target point, the external temperature value, the duration of the bonding process, the type of the first silicone pad (110) and the type of the second silicone pad (210); Determining a control amount corresponding to the target point at the predicted moment according to the predicted position; The opening and closing state of the electric control valve (310) corresponding to the target point and the real-time power of the working pump (500) are controlled according to the control amount.

9. The method for applying a flexible curved surface film according to claim 8, wherein: The method further comprises: determining a target position compensation value according to the predicted position and a theoretical position corresponding to the target point at the predicted moment; determining a compensation power value of the working pump (500) according to the target position compensation value; The real-time power of the working pump (500) is adjusted according to the power compensation value.

Images