Control device, winding device, construction method of control system, control system, control method, and storage medium

Abstract

The invention provides a control device, a winding device, a construction method of a control system, the control system, a control method and a storage medium, which can make a control amount from a control object closer to a target value. A control device according to an embodiment controls a take-up mechanism capable of taking up an object. The control device includes a first control unit, a first calculation unit, and a second calculation unit. The first control unit acquires, from the winding mechanism, a control amount to be controlled. The first control unit acquires a first operation amount corresponding to an operation amount when the control amount is output from the take-up mechanism by inputting the control amount into a model that has been subjected to machine learning in advance. The first calculation unit calculates a difference between the first operation amount and a second operation amount sent toward the winding mechanism. The second calculation unit inputs a third operation amount obtained by adding the difference to the second operation amount to the winding mechanism.

Description

Technical Field

[0001] Embodiments of the present invention relate to a control device, a winding device, a construction method of a control system, a control system, a control method, and a storage medium. Background Art

[0002] There is a device for controlling a controlled object. Regarding this control device, a technology capable of bringing a controlled amount from the controlled object closer to a target value is required.

[0003] Patent Document 1: Japanese Patent No. 6813231 Summary of the invention

[0004] The problem to be solved by the present invention is to provide a control device, a winding device, a control system construction method, a control system, a control method and a storage medium, which can make the control amount from the control object closer to the target value.

[0005] The control device of the embodiment controls a winding mechanism capable of winding an object. The control device includes a first control unit, a first operation unit, and a second operation unit. The first control unit obtains a control amount as a controlled object from the winding mechanism. The first control unit inputs the control amount to a model that has been machine-learned in advance, thereby obtaining a first operation amount that is equivalent to the operation amount when the control amount is output from the winding mechanism. The first operation unit calculates the difference between the first operation amount and the second operation amount sent to the winding mechanism. The second operation unit inputs a third operation amount obtained by adding the difference to the second operation amount to the winding mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 It is a block diagram showing the structure of the winding device according to the embodiment.

[0007] Figure 2 It is a perspective view showing a winding device according to the embodiment.

[0008] Figure 3 It is a block diagram showing the structure of a winding device 1r of a reference example.

[0009] Figure 4 (a) Figure 4 (b) is data representing the characteristics of the control system of the embodiment.

[0010] Figure 5 It is a block diagram showing the configuration of a winding device according to a modified example of the embodiment.

[0011] Figure 6 This is a schematic diagram showing the hardware configuration. DETAILED DESCRIPTION

[0012] Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. The accompanying drawings are schematic or conceptual, and the relationship between the thickness and width of each part, the ratio of the size between the parts, etc. may not be the same as in reality. Even when representing the same part, the size and ratio of each other are sometimes represented differently according to the accompanying drawings. In the specification and each figure of this application, the same symbols are marked for the same elements as those already described, and the detailed description is appropriately omitted.

[0013] Figure 1 It is a block diagram showing the structure of the winding device according to the embodiment.

[0014] like Figure 1 As shown, the winding device 1 of the embodiment includes a control device 10 and a winding mechanism 20. The winding mechanism 20 can wind up an object. The object is a foil, a wire, etc. The control device 10 controls the operation of the winding mechanism 20.

[0015] The control device 10 includes a first control unit 11, a second control unit 12, a first operation unit 13, a second operation unit 14 and an operation unit 15. The first control unit 11 obtains a control quantity y (output value) from the winding mechanism 20. "Control quantity" refers to a quantity that becomes a controlled object among quantities belonging to a control object. When the first control unit 11 obtains the control quantity y, it refers to a model 16. The model 16 performs machine learning in advance so that when a control quantity is input, it estimates an operation quantity corresponding to the control quantity. "Operation quantity" refers to a quantity applied to a control object in order to control the control quantity in a control system. Regarding the definitions of control quantity and operation quantity, the contents defined in JIS Z8116-1994 can be used.

[0016] For example, model 16 includes a decision tree or a neural network. Model 16 may include a random forest as a decision tree. In the case of using a neural network, model 16 preferably includes a recurrent neural network (RNN). RNN may also include a long short-term memory network (Long Short-Term Memory) structure.

[0017] The first control unit 11 inputs the control amount y into the model 16, and obtains the first operation amount u1 estimated by the model 16. The first operation amount u1 is an operation amount corresponding to the control amount y. The control amount y output from the winding mechanism 20 is equivalent to the control amount when the first operation amount u1 is input to the winding mechanism 20. In other words, the first operation amount u1 is equivalent to the operation amount when the control amount y is output from the winding mechanism 20.

[0018] The operation unit 15 receives the target value r of the action of the winding mechanism 20 from the upper system. "Target value" refers to a quantity given as a target in the control system in order to make the controlled quantity take the value. Regarding the definition of the target value, the content defined in JISZ8116-1994 can also be used. The operation unit 15 calculates the error e between the target value r and the controlled quantity y. The second control unit 12 calculates the second operation amount u2 to compensate for the error e and sends it to the winding mechanism 20.

[0019] When the first control unit 11 estimates the first manipulated variable u1, the target value r may be used in addition to the controlled variable y. In this case, the model 16 performs machine learning in advance so as to output the manipulated variable according to the input of the controlled variable and the target value. The first control unit 11 inputs the controlled variable y and the target value r into the model 16, thereby obtaining the first manipulated variable u1.

[0020] The first calculation unit 13 obtains the first operation amount u1 and the second operation amount u2, and calculates a difference d1 between the first operation amount u1 and the second operation amount u2. The difference d1 corresponds to a disturbance estimated based on the control amount from the winding mechanism 20.

[0021] The second operation unit 14 calculates the third operation amount u3 by adding the difference d1 to the second operation amount u2. The second operation unit 14 inputs the third operation amount u3 to the winding mechanism 20. The winding mechanism 20 operates according to the third operation amount u3.

[0022] When the disturbance d2 is applied to the winding mechanism 20, the control amount y changes according to the disturbance d2. The first operation amount u1 also changes according to the change in the control amount y, and as a result, the third operation amount u3 also changes.

[0023] As a specific example, the control amount y decreases due to the disturbance d2. In accordance with the decrease in the control amount y, the first operation amount u1 also decreases. The difference d1 between the first operation amount u1 and the second operation amount u2 increases. The difference d1 is added to the second operation amount u2. Thus, the operation amount of the winding mechanism 20 is corrected according to the disturbance d2 to output the target value.

[0024] Figure 2 It is a perspective view showing a winding device according to the embodiment.

[0025] For example, Figure 2 As shown in FIG. 1 , the winding mechanism 20 includes a first roller 21, a second roller 22, an intermediate roller 23, a sensor 24, a movable portion 25, and a movable portion 26. Figure 2 In the example shown, the object O is a foil (film).

[0026] The first roller 21 winds up the object O by rotating. Specifically, the first roller 21 includes a frame 21a, a rotating shaft 21b, and a motor 21c. The frame 21a is a component for winding the object O. In the example shown in the figure, the frame 21a is circular, but the specific shape of the frame 21a is arbitrary. The rotating shaft 21b is installed at the center of the frame 21a. The motor 21c rotates the rotating shaft 21b. The frame 21a rotates around the rotating shaft 21b by the action of the motor 21c.

[0027] The object O is pre-wound on the second roller 22. The second roller 22 delivers the object O to the first roller 21 by rotating. The second roller 22 can apply the same structure as the first roller 21. For example, the second roller 22 includes a frame 22a, a rotating shaft 22b and a motor 22c. The frame 22a is a component pre-wound with the object O. The frame 22a can be circular or have other shapes. The rotating shaft 22b is installed at the center of the frame 22a. In the example shown in the figure, the rotating shaft 21b and the rotating shaft 22b are parallel to each other. The motor 22c rotates the rotating shaft 22b. Through the action of the motor 22c, the frame 22a rotates around the rotating shaft 22b.

[0028] Here, for convenience of description, a direction parallel to the rotation axis 21b or the rotation axis 22b is referred to as an "X direction". A direction intersecting the X direction is referred to as a "Y direction". A direction intersecting the XY plane is referred to as a "Z direction".

[0029] While the object O is passing from the first roller 21 to the second roller 22, the object O contacts the plurality of intermediate rollers 23. Each intermediate roller 23 is rotatable about the X direction. Each intermediate roller 23 changes the conveying direction of the object O and applies appropriate tension to the object O. In the example shown in the figure, ten intermediate rollers 23, namely, intermediate rollers 23a to 23j, are provided in the section where the object O is passing from the first roller 21 to the second roller 22.

[0030] The sensor 24 detects parameters related to the object O. For example, the sensor 24 includes an encoder 24a, a force sensor 24b, a force sensor 24c, a light sensor 24d, a light sensor 24e, and a camera (image sensor) 24f.

[0031] The encoder 24a detects the rotation speed of the intermediate roller 23a. Specifically, the encoder 24a is an optical or magnetic encoder, and counts the optical signal or magnetic signal detected by the rotation of the intermediate roller 23a. The encoder 24a calculates the rotation speed of the intermediate roller 23a based on the rotation speed of the intermediate roller 23a per unit time. The rotation speed of the intermediate roller 23a is equivalent to the conveying speed of the object O.

[0032] The force sensor 24b is attached to the intermediate roller 23b. A force corresponding to the tension of the object O is applied to the intermediate roller 23b in a direction intersecting the X direction. The force sensor 24b detects the force applied to the intermediate roller 23b as the tension of the object O.

[0033] The force sensor 24c is attached to the intermediate roller 23i. A force corresponding to the tension of the object O is applied to the intermediate roller 23i in a direction intersecting the X direction. The force sensor 24c detects the force applied to the intermediate roller 23i as the tension of the object O.

[0034] The optical sensor 24d and the optical sensor 24e detect the position of the object O in the direction perpendicular to the conveying direction. In the example shown in the figure, the optical sensor 24d detects the position of the object O in the X direction between the intermediate rollers 23c and 23d. The optical sensor 24e detects the position of the object O in the X direction between the intermediate rollers 23g and 23h.

[0035] The camera 24f detects the positional offset of the object O in the direction perpendicular to the conveying direction and the angular offset of the conveying direction of the object O. For example, based on the image obtained by the camera 24f, the end of the foil in the perpendicular direction is detected. The distance between the position of the end in the image and the preset position is the positional offset. In addition, the angle between the preset line and the line representing the end (edge) of the foil is the angular offset. The preset line is set to be parallel to the direction connecting the adjacent rollers. In the example shown in the figure, the camera 24f detects the positional offset and the tilt of the object O between the intermediate roller 23h and the intermediate roller 23i.

[0036] The movable part 25 includes an actuator and can drive the first roller 21 in the X direction. The position of the object O wound on the first roller 21 in the X direction changes due to the operation of the movable part 25. In the example shown in the figure, the movable part 25 includes a base frame 25a and a stage 25b. The base frame 25a is a plate-shaped member parallel to the YZ plane. The stage 25b is a plate-shaped member parallel to the XY plane. The stage 25b is movable in the X direction and supports the base frame 25a from below. The first roller 21, the intermediate rollers 23f to 23j, the force sensor 24c, the optical sensor 24e and the camera 24f are mounted on the base frame 25a. As the stage 25b moves in the X direction, the position of each element mounted on the base frame 25a in the X direction changes.

[0037] The movable part 26 includes an actuator and can drive the second roller 22 in the X direction. The position of the object O sent out from the second roller 22 in the X direction changes due to the operation of the movable part 26. The movable part 26 can apply the same structure as the movable part 25. For example, the movable part 26 includes a base frame 26a and a stage 26b. The base frame 26a is a plate-shaped member parallel to the YZ plane. The stage 26b is a plate-shaped member parallel to the XY plane. The stage 26b can move in the X direction and supports the base frame 26a from below. The second roller 22, the intermediate rollers 23a to 23e, the encoder 24a, the force sensor 24b, and the optical sensor 24d are mounted on the base frame 26a. As the stage 26b moves in the X direction, the position of each element mounted on the base frame 26a in the X direction changes.

[0038] Figure 2 The winding mechanism 20 shown outputs the values ​​detected by the sensors 24a to 24f. Figure 1 The control amount y shown includes the conveying speed detected by the encoder 24a, the tension detected by the force sensor 24b, the tension detected by the force sensor 24c, the position detected by the optical sensor 24d, the position detected by the optical sensor 24e, and the position offset and tilt detected by the camera 24f. The control device 10 receives these control amounts.

[0039] In addition, the control device 10 calculates the operation amounts of the motor 21c, the motor 22c, the movable portion 25, and the movable portion 26. That is, the first operation amount u1 to the third operation amount u3 include the rotation speed of the motor 21c, the rotation speed of the motor 22c, the position of the first roller 21 (movable portion 25), and the position of the second roller 22 (movable portion 26).

[0040] The model 16 used for estimating the first manipulated variable u1 is preliminarily subjected to machine learning. In machine learning, a plurality of learning data sets are used. Each learning data set includes a controlled variable and a manipulated variable. The model 16 repeatedly learns to output the manipulated variable according to the input of the controlled variable. For example, a learning data set is collected by causing the controlled object to move in a state where there is little or substantially no interference with the controlled object. The model 16 is subjected to learning using the collected learning data set.

[0041] One advantage of the embodiment will be described.

[0042] During the operation of the winding mechanism 20, disturbance d2 may be generated due to various reasons. For example, disturbance d2 may be generated due to changes in the material of the object O, partial damage to the object O, twisting of the object O, or changes in the winding mechanism 20 over time. If the disturbance d2 is generated, the conveying speed of the object O, the tension of the object O, or the position of the object O may change. If these values ​​change, the behavior of the object O may deviate from the desired state, resulting in wrinkles, twists, or breaks in the foil.

[0043] Figure 3 It is a block diagram showing the structure of a winding device 1r of a reference example.

[0044] Figure 3 The winding device 1r shown includes a control device 10r instead of the control device 10. The control device 10r does not include the first control unit 11, the first operation unit 13, and the second operation unit 14 compared to the control device 10. In the winding device 1r, the second operation amount u2 corresponding to the target value r and the control amount y is sent from the second control unit 12 to the winding mechanism 20. Thus, the winding mechanism 20 is controlled so that the difference between the control amount y and the target value r becomes smaller. However, in the case of a disturbance d2, even if the second operation amount u2 is input to the winding mechanism 20, it may be difficult to make the control amount y close to the target value r.

[0045] In order to make the control amount y closer to the target value r, it is effective to estimate the first operation amount u1 corresponding to the actual input to the winding mechanism 20 based on the control amount y. By estimating the first operation amount u1, the second operation amount u2 can be corrected and the third operation amount u3 can be calculated. On the other hand, two or more operation amounts such as the position of the first roller 21 (movable part 25), the position of the second roller 22 (movable part 26), the rotation speed of the first roller 21, and the rotation speed of the second roller 22 are input to the winding mechanism 20. In addition, two or more operation amounts such as the conveying speed of the object O, the tension of the object O, and the position of the object O are output from the winding mechanism 20. There is a nonlinear relationship between the multiple operation amounts input to the winding mechanism 20 and the multiple control amounts output from the winding mechanism 20. Therefore, it is essentially impossible to analyze the relationship between the control amount y and the first operation amount u1 by regression analysis or the like. Therefore, with respect to the winding mechanism 20, the control amount y is adjusted by inputting the second operation amount u2 corresponding to the error e into the winding mechanism 20.

[0046] Regarding this subject, the control device 10 of the embodiment includes a first control unit 11, a first operation unit 13, and a second operation unit 14. The first control unit 11 uses a model 16 to estimate a first operation amount u1 corresponding to the control amount y. The model 16 includes a decision tree or a neural network, and machine learning is performed in advance. By using the model 16, the first operation amount u1 having a nonlinear relationship with the control amount y can be estimated based on the control amount y. By obtaining the first operation amount u1, the difference d1 corresponding to the disturbance d2 can be calculated. The third operation amount u3 can be calculated using the difference d1. As a result, the control amount y can be made closer to the target value r.

[0047] Another advantage of the embodiment will be described.

[0048] The control device 10 of the preferred embodiment can also be introduced to the control object after the operation starts. Figure 3 After the operation of the winding device 1r shown, the first control unit 11, the first operation unit 13, and the second operation unit 14 can be installed on the winding device 1r. Regarding the learning of the model 16, the operation amount of the winding mechanism 20 and the control amount from the winding mechanism 20 are collected in the state where the operation of the winding mechanism 20 has started. The collected data is used as a learning data set for the learning of the model 16. After the learning of the model 16, the first control unit 11, the first operation unit 13, and the second operation unit 14 are added to the control device 10r, thereby realizing the control device 10 of the embodiment.

[0049] That is, according to an embodiment of the present invention, Figure 1 As shown, a control device 10b (second control device) including the first control unit 11, the first operation unit 13, the second operation unit 14, and the model 16 may be added to the control device 10a (first control device) including the second control unit 12 and the model 16. By adding the new control device 10b to the control system including the existing control device 10a, a new control system can be constructed.

[0050] The embodiment of the present invention can also be applied to control objects other than winding devices. That is, the control object of the control device 10 is arbitrary. The first control unit 11 obtains the control amount from the control object and inputs the control amount into the model 16, thereby obtaining the first operation amount u1. The first operation unit 13 calculates the difference d1 between the first operation amount u1 and the second operation amount u2 sent to the control object. The second operation unit 14 inputs the third operation amount u3 obtained by adding the difference d1 to the second operation amount u2 into the control object. Thereby, the control amount y of the control object can be made closer to the target value r. In addition, by adding the control device 10b to the existing device, the control device 10 of the embodiment can be realized.

[0051] The embodiment is particularly effective when the operation amount and the control amount have a nonlinear relationship. When the operation amount and the control amount have a nonlinear relationship, it is difficult to perform regression analysis manually. By using the model 16, the operation amount can be estimated based on the control amount with high accuracy.

[0052] More preferably, the model 16 includes an RNN, and the first control unit 11 inputs the time series data of the controlled amount into the model 16. The inventors of the present application have verified that when the manipulated amount and the controlled amount have a nonlinear relationship, the error of the estimated manipulated amount can be reduced by using the RNN and the time series data.

[0053] Figure 4 (a) and Figure 4 (b) is data representing the characteristics of the control system of the embodiment.

[0054] Figure 4 (a) shows the characteristics when the model includes a neural network other than RNN. Figure 4 (b) shows the characteristics when the model includes RNN. Figure 4 (a) and Figure 4 In the graph of (b), the horizontal axis represents the estimated number of executions. The vertical axis represents one of the operation amounts estimated by the model. The solid line represents the operation amount estimated by the model, and the dotted line represents the operation amount of the correct answer. The control amount input to the model includes the conveying speed detected by the encoder 24a, the tension detected by the force sensor 24b, the tension detected by the force sensor 24c, the position detected by the optical sensor 24d, the position detected by the optical sensor 24e, and the position offset and tilt detected by the camera 24f. The operation amount estimated by the model includes the position of the first roller 21, the position of the second roller 22, the rotation speed of the first roller 21, and the rotation speed of the second roller 22.

[0055] from Figure 4 The results shown in (a) show that the model including the neural network can roughly estimate the operation amount close to the correct answer. The root mean square error (RMSE) between the estimated value and the correct answer is 70.010. The mean absolute error (MAE) is 46.833. These results confirm that the model including the neural network can estimate the operation amount with high accuracy.

[0056] from Figure 4 The results shown in (b) show that the model including RNN can estimate the operation amount with higher accuracy than the model including general neural network. Specifically, when the model including RNN is used, the RMSE is 56.164. The MAE is 46.512. Regarding RMSE, a large improvement of about 20% in accuracy was confirmed.

[0057] According to the result, it is preferred that the model includes RNN and the time series data is input into the model. Thus, the accuracy of the estimated operation amount can be further improved. By improving the accuracy of the estimated operation amount, the accuracy of the difference d1 equivalent to the disturbance is also improved. As a result, the control amount output from the control object can be made closer to the target value.

[0058] (Variation Example)

[0059] Figure 5 It is a block diagram showing the configuration of a winding device according to a modified example of the embodiment.

[0060] Figure 5 The winding device 2 shown includes a control device 10x and a winding mechanism 20. The control device 10x further includes a determination unit 17 compared to the control device 10. The functions of the elements in the control device 10x other than the determination unit 17 are the same as those of the elements in the control device 10.

[0061] The determination unit 17 receives the difference d1 from the first operation unit 13. In addition, the determination unit 17 sends the difference d1 to the second operation unit 14. When receiving the difference d1, the determination unit 17 determines whether the winding mechanism 20 is normal based on the difference d1. For example, the determination unit 17 compares the difference d1 with a preset threshold value. When the difference d1 is greater than the threshold value, the determination unit 17 determines that the winding mechanism 20 is abnormal.

[0062] When it is determined that the winding mechanism 20 is abnormal, the determination unit 17 outputs the determination result. For example, the determination unit 17 sends the determination result to a preset terminal device. When the winding mechanism 20 is equipped with an output device (speaker, light or monitor), the determination unit 17 may also cause the output device to output an output indicating the determination result. When it is determined that the winding mechanism 20 is abnormal, the determination unit 17 may also stop the operation of the winding mechanism 20.

[0063] If the difference d1 is too large, there is a possibility that an abnormality has occurred in the winding mechanism 20 or the object O. By providing the determination unit 17, an abnormality in the winding device 2 can be detected earlier. For example, it is possible to avoid continuing to process the object O in an undesirable state.

[0064] Figure 6 This is a schematic diagram showing the hardware configuration.

[0065] As the first control unit 11, the second control unit 12 and the determination unit 17, for example, Figure 6 The computer 90 shown in FIG. 9 includes a CPU 91 , a ROM 92 , a RAM 93 , a storage device 94 , an input interface 95 , an output interface 96 , and a communication interface 97 .

[0066] The ROM 92 stores a program for controlling the operation of the computer 90. The ROM 92 stores a program necessary for the computer 90 to realize the above-mentioned processing. The RAM 93 functions as a storage area for developing the program stored in the ROM 92.

[0067] The CPU 91 includes a processing circuit and uses the RAM 93 as a working memory to execute a program stored in at least one of the ROM 92 and the storage device 94. While executing the program, the CPU 91 controls each component via the system bus 98 and executes various processes.

[0068] The storage device 94 stores data necessary for executing the program and data obtained by executing the program.

[0069] An input interface (I / F) 95 can connect the computer 90 to an input device 95a. The input I / F 95 is, for example, a serial bus interface such as USB. The CPU 91 can read various data from the input device 95a via the input I / F 95.

[0070] The output interface (I / F) 96 can connect the computer 90 to the output device 96a. The output I / F 96 is, for example, a video output interface such as a digital visual interface (DVI) or a high-definition multimedia interface (HPMI (registered trademark)). The CPU 91 can send data to the output device 96a via the output I / F 96 to cause the output device 96a to display an image.

[0071] The communication interface (I / F) 97 can connect a server 97a outside the computer 90 to the computer 90. The communication I / F 97 is, for example, a network card such as a LAN card. The CPU 91 can read various data from the server 97a via the communication I / F 97.

[0072] The storage device 94 includes one or more selected from a hard disk drive (HDD) and a solid state drive (SSD). The input device 95a includes one or more selected from a mouse, a keyboard, a microphone (sound input), and a touch panel. The output device 96a includes one or more selected from a monitor, a projector, a printer, and a speaker. A device having the functions of both the input device 95a and the output device 96a, such as a touch panel, may also be used.

[0073] Each process executed by the first control unit 11, the second control unit 12, and the determination unit 17 may be implemented by a single computer 90 or by cooperation of a plurality of computers 90. For example, a computer 90 functioning as the second control unit 12 and the determination unit 17 may be added to the control system including the computer 90 functioning as the first control unit 11.

[0074] As the first operation unit 13, the second operation unit 14, and the operation unit 15, for example, a general-purpose operation circuit inside a programmable logic controller (PLC) is used. As the first operation unit 13, the second operation unit 14, and the operation unit 15, an electric operation circuit may also be used. The computer 90 that functions as the first control unit 11 or the second control unit 12 may also have the functions of the first operation unit 13, the second operation unit 14, or the operation unit 15.

[0075] The processing of the above-mentioned various data is recorded as a program that can be executed by a computer on a disk (floppy disk, hard disk, etc.), an optical disk (CD-ROM, CD-R, CD-RW, DVD-ROM, DVD±R, DVD±RW, etc.), a semiconductor memory or other non-transitory computer-readable storage medium.

[0076] For example, the information recorded in the recording medium can be read out by a computer (or embedded system). In the recording medium, the recording format (storage format) is arbitrary. For example, the computer reads out the program from the recording medium and makes the CPU execute the instructions described in the program based on the program. In the computer, the acquisition (or reading) of the program can also be carried out through a network.

[0077] Embodiments of the present invention include the following features.

[0078] (Feature 1)

[0079] A control device controls a winding mechanism capable of winding an object, comprising:

[0080] a first control unit that obtains a control amount as a controlled object from the winding mechanism, inputs the control amount into a model that has been subjected to machine learning in advance, and thereby obtains a first operation amount that corresponds to an operation amount when the control amount is output from the winding mechanism;

[0081] a first calculation unit that calculates a difference between the first operation amount and a second operation amount sent to the winding mechanism; and

[0082] The second calculation unit inputs a third operation amount obtained by adding the difference to the second operation amount into the winding mechanism.

[0083] (Feature 2)

[0084] In the control device according to feature 1,

[0085] The winding mechanism includes a first roller for winding the object.

[0086] The first operation amount, the second operation amount, and the third operation amount include a rotation speed of the first roller.

[0087] (Feature 3)

[0088] In the control device according to feature 1 or feature 2,

[0089] The winding mechanism includes a first roller for winding the object.

[0090] The first roller rotates around the axial direction.

[0091] The first operation amount, the second operation amount, and the third operation amount include a position of the first roller in the axial direction.

[0092] (Feature 4)

[0093] In the control device described in any one of features 1 to 3,

[0094] The winding mechanism includes a second roller for delivering the object.

[0095] The first operation amount, the second operation amount, and the third operation amount include a rotation speed of the second roller.

[0096] (Feature 5)

[0097] In the control device according to any one of features 1 to 4,

[0098] The control amount includes two or more selected from the tension of the object, the conveyance speed of the object, and the position of the object in a direction perpendicular to the conveyance direction of the object.

[0099] (Feature 6)

[0100] In the control device according to any one of features 1 to 5,

[0101] The above models include decision trees or neural networks.

[0102] (Feature 7)

[0103] In the control device according to any one of features 1 to 5,

[0104] The above model contains a recurrent neural network,

[0105] The first control unit inputs the time series data of the control amount into the model.

[0106] (Feature 8)

[0107] In the control device described in any one of features 1 to 7,

[0108] The control device further includes a second control unit that calculates the second operation amount based on a target value for the winding mechanism.

[0109] (Feature 9)

[0110] In the control device according to any one of features 1 to 8,

[0111] The control device further includes a determination unit that determines whether the winding mechanism is normal based on the first operation amount.

[0112] (Feature 10)

[0113] A winding device, comprising:

[0114] The control device according to any one of features 1 to 9; and

[0115] The above-mentioned winding mechanism.

[0116] According to the above-described embodiments, a control device, a winding device, a control system construction method, a control system, a control method, a program, and a storage medium can be provided, which can make the controlled amount from the controlled object closer to the target value.

[0117] In the present specification, "or" means that "at least one or more" of the items listed in the article can be adopted.

[0118] Several embodiments of the present invention are illustrated above, but these embodiments are presented as examples and are not intended to limit the scope of the invention. These new embodiments can be implemented in various other ways, and various omissions, substitutions, changes, etc. can be made without departing from the scope of the main purpose of the invention. These embodiments and their variations are included in the scope and main purpose of the invention, and are included in the invention described in the technical solution and the scope equivalent thereto. Moreover, the above-mentioned embodiments can be combined with each other and implemented.

Claims

1. A control device for controlling a winding mechanism capable of winding an object, comprising: a first control unit that obtains a control amount as a controlled object from the winding mechanism, inputs the control amount into a model that has been subjected to machine learning in advance, and thereby obtains a first operation amount that corresponds to an operation amount when the control amount is output from the winding mechanism; a first calculation unit that calculates a difference between the first operation amount and a second operation amount sent to the winding mechanism; and The second calculation unit inputs a third operation amount obtained by adding the difference to the second operation amount into the winding mechanism.

2. The control device according to claim 1, wherein: The winding mechanism includes a first roller for winding the object. The first operation amount, the second operation amount, and the third operation amount include a rotation speed of the first roller.

3. The control device according to claim 1 or 2, wherein: The winding mechanism includes a first roller for winding the object. The first roller rotates around the axial direction. The first operation amount, the second operation amount, and the third operation amount include a position of the first roller in the axial direction.

4. The control device according to claim 1 or 2, wherein: The winding mechanism includes a second roller for delivering the object. The first operation amount, the second operation amount, and the third operation amount include a rotation speed of the second roller.

5. The control device according to claim 1 or 2, wherein: The control amount includes two or more selected from the tension of the object, the conveyance speed of the object, and the position of the object in a direction perpendicular to the conveyance direction of the object.

6. The control device according to claim 1 or 2, wherein: The above models include decision trees or neural networks.

7. The control device according to claim 1 or 2, wherein: The above model contains a recurrent neural network, The first control unit inputs the time series data of the control amount into the model.

8. The control device according to claim 1 or 2, wherein: The control device further includes a second control unit that calculates the second operation amount based on a target value for the winding mechanism.

9. The control device according to claim 1 or 2, wherein: The control device further includes a determination unit that determines whether the winding mechanism is normal based on the first operation amount.

10. A winding device comprising: The control device according to claim 1 or 2; and The above-mentioned winding mechanism.

11. A method for constructing a control system, wherein: A second control device is added to a control system including a first control device and a control object controlled by the first control device. The second control device comprises: A first control unit that obtains a control amount as a controlled object from the control object, and inputs time series data of the control amount into a model that has been pre-machined to obtain a first operation amount that corresponds to an operation amount when the control amount is output from the control object; a first operation unit that calculates a difference between the first operation amount and a second operation amount sent from the control device to the control object; and The second operation unit inputs a third operation amount obtained by adding the difference to the second operation amount into the control object.

12. The method for constructing a control system according to claim 11, wherein: The above model contains a recurrent neural network.

13. A control device for controlling a controlled object, comprising: A first control unit that obtains a control amount as a controlled object from the control object, and inputs time series data of the control amount into a model that has been pre-machined to obtain a first operation amount that corresponds to an operation amount when the control amount is output from the control object; a first operation unit that calculates a difference between the first operation amount and a second operation amount sent to the control object; and The second operation unit inputs a third operation amount obtained by adding the difference to the second operation amount into the control object.

14. A control system comprising: The control device of claim 13; and The above control object.

15. A control method for controlling a winding mechanism capable of winding an object, wherein: A control amount as a controlled object is obtained from the winding mechanism, and the control amount is input to a model that has been subjected to machine learning in advance, thereby obtaining a first operation amount that corresponds to an operation amount when the control amount is output from the winding mechanism, calculating the difference between the first operation amount and the second operation amount sent to the winding mechanism, A third operation amount obtained by adding the difference to the second operation amount is input to the winding mechanism.

16. A storage medium, wherein: A program for causing a computer to execute the control method according to claim 15 is stored.

Images