Material correction system, material correction method, program and integration system

The material correction system addresses inaccuracies in estimating warpage post-correction by employing staggered roll groups, sensors, and a learning model to enhance estimation precision, effectively flattening materials for subsequent processing.

JP2025011756APending Publication Date: 2025-01-24PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD

Patent Information

Application Number
JP2023114058
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-07-11
Publication Date
2025-01-24

AI Technical Summary

Technical Problem

Existing methods for estimating the deformation status of orthodontic materials during correction suffer from inaccuracies due to the limitations in the estimation of material constants and expression parameters, leading to decreased accuracy in predicting warpage after correction.

Method used

A material correction system utilizing opposing roll groups with staggered arrangements of multiple rolls, equipped with an acquisition unit to gather parameters from sensors, and an estimation unit to estimate warpage using a learning model, and a learning unit to refine the model based on actual measurements.

Benefits of technology

Accurately estimates and corrects warpage in plate materials by using a learning model that improves estimation accuracy through iterative learning, ensuring precise flattening of materials for further processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2025011756000001_ABST
    Figure 2025011756000001_ABST
Patent Text Reader

Abstract

To accurately estimate a warpage amount after correction of a plate material which is a correction object.SOLUTION: A material correction system 1 penetrates a plate material between facing roll groups R1 and R2 and corrects warpage of a plate material. The material correction system 1 includes an acquisition part 101, and an estimation part 102. The acquisition part 101 acquires information on a parameter using at least one of an initial warpage amount which is a warpage amount of a plate material before correction, a plate thickness of the plate material, a position of a first correction roll group which is one roll group out of the facing roll groups, and a roll load applied to the first correction roll group R1, as the parameter, on the basis of a detection result of a sensor 20 for detecting the information. The estimation part 102 estimates a warpage amount after correction, using a learning model estimating the warpage amount after correction with the parameter acquired by the acquisition part 101 as an explanatory variable, and the warpage amount after correction which is a warpage amount of the plate material after correction as an objective variable.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure generally relates to a material straightening system, a material straightening method, a program, and an integrated system, and more particularly to a material straightening system, a material straightening method, a program, and an integrated system for straightening a warp of a plate material. [Background technology]

[0002] Patent Document 1 describes a roller leveler that accurately estimates the deformation state and material constants of a workpiece being straightened by roller straightening. The roller leveler in Patent Document 1 obtains the relationship between the material constants of the workpiece being straightened and any two expression parameters expressing the deformation state of the workpiece being straightened, and the entry-side straightening load, exit-side straightening load, and total straightening gravity generated in the roller leveler, estimates the material constants and the two expression parameters of the workpiece being straightened from the above relationship based on the entry-side straightening load, exit-side straightening load, and total straightening power measured under load, and estimates the deformation state of the workpiece being straightened based on the estimated material constants and the two expression parameters of the workpiece being straightened. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-175443 A Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, the deformation state of the material to be straightened during straightening is estimated based on the estimated material constants and two expression parameters of the material to be straightened. Therefore, depending on the estimation accuracy of the material constants and the two expression parameters, the estimation accuracy of the deformation state (amount of warping) of the material to be straightened (plate material) may decrease.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and aims to provide a material straightening system, a material straightening method, a program, and an integrated system that can accurately estimate the amount of warpage after straightening of a plate material to be straightened. [Means for solving the problem]

[0006] In a material straightening system according to an embodiment of the present disclosure, roll groups including a plurality of rolls face each other, and a plate material is passed between the facing roll groups to straighten the warp of the plate material. A plurality of rolls included in one of the facing roll groups and a plurality of rolls included in the other roll group are arranged in a staggered pattern. The material straightening system includes an acquisition unit and an estimation unit. The acquisition unit acquires at least one of an initial warp amount, which is the amount of warp of the plate material before straightening, a plate material thickness, a position of a first straightening roll group, which is one of the facing roll groups, and a roll load applied to the first straightening roll group, as parameters based on a detection result of a sensor that detects information related to the parameters. The estimation unit estimates the warp amount after straightening using a learning model that estimates the warp amount after straightening, which is the amount of warp of the plate material after straightening, as a target variable, using the parameters acquired by the acquisition unit as explanatory variables.

[0007] In a material straightening system according to an embodiment of the present disclosure, roll groups including a plurality of rolls face each other, and a plate material is passed between the facing roll groups to straighten the warp of the plate material. A plurality of rolls included in one of the facing roll groups and a plurality of rolls included in the other roll group are arranged in a staggered pattern. The material straightening system includes an acquisition unit and a learning unit. The acquisition unit acquires at least one of an initial warp amount, which is the amount of warp of the plate material before straightening, a plate thickness of the plate material, a position of a first straightening roll group, which is one of the facing roll groups, and a roll load applied to the first straightening roll group, as parameters based on a detection result of a sensor that detects information related to the parameters. The acquisition unit acquires an actual warp amount, which is the amount of warp measured for the plate material after straightening. The learning unit performs learning using the parameters acquired by the acquisition unit as explanatory variables and the actual warp amount as a target variable to generate a learning model.

[0008] A material straightening method according to one aspect of the present disclosure is used in a material straightening system. The material straightening system is a system in which roll groups including a plurality of rolls face each other, and a plate material is passed between the facing roll groups to straighten the warp of the plate material. A plurality of rolls included in one of the facing roll groups and a plurality of rolls included in the other roll group are arranged in a staggered pattern. The material straightening method includes an acquisition step and an estimation step. The acquisition step acquires at least one of an initial warp amount, which is the warp amount of the plate material before straightening, a plate material thickness, a position of a first straightening roll group, which is one of the facing roll groups, and a roll load applied to the first straightening roll group, as a parameter based on a detection result of a sensor that detects information related to the parameter. The estimation step estimates the warp amount after straightening using a learning model that estimates the parameter acquired in the acquisition step as an explanatory variable and a post-straightening warp amount, which is the warp amount of the plate material after straightening, as an objective variable.

[0009] A material straightening method according to one aspect of the present disclosure is used in a material straightening system. The material straightening system is a system in which roll groups including a plurality of rolls face each other, and a plate material is passed between the facing roll groups to straighten the warp of the plate material. A plurality of rolls included in one of the facing roll groups and a plurality of rolls included in the other roll group are arranged in a staggered pattern. The material straightening method includes an acquisition step and a learning step. The acquisition step acquires at least one of an initial warp amount, which is the amount of warp of the plate material before straightening, a plate material thickness, a position of a first straightening roll group, which is one of the facing roll groups, and a roll load applied to the first straightening roll group, as a parameter based on a detection result of a sensor that detects information related to the parameter. The acquisition step acquires an actual warp amount, which is the amount of warp measured for the plate material after straightening. The learning step performs learning using the parameters acquired in the acquisition step as explanatory variables and the actual warp amount acquired in the acquisition step as a target variable, to generate a learning model.

[0010] A program according to one aspect of the present disclosure is a program for causing a computer system to execute the material straightening method.

[0011] An integrated system according to one aspect of the present disclosure includes the material straightening system and a sensor that detects information related to the parameter acquired by the material straightening system. Effect of the Invention

[0012] According to the present disclosure, it is possible to accurately estimate the amount of warpage of a plate material to be straightened after the straightening. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing the configuration of a material straightening device as a material straightening system according to one embodiment, and an integrated system including the material straightening device. [Diagram 2] FIG. 2 is a system configuration diagram showing a usage form of the integrated system. [Diagram 3] FIG. 3 is a schematic diagram illustrating the configuration of the integrated system. [Figure 4] FIG. 4 is a flow chart showing the operation of the material straightening apparatus in the inference phase. [Diagram 5] FIG. 5 is a flow chart showing the operation of the material straightening apparatus in the learning phase. [Figure 6] FIG. 6 is a block diagram showing a configuration of a material straightening device as a material straightening system according to a first modified example of an embodiment, and an integrated system including the material straightening device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] The embodiment and modified examples described below are merely examples of the present disclosure, and the present disclosure is not limited to the embodiment and modified examples. Various modifications other than the embodiment and modified examples are possible according to the design, etc., as long as they do not deviate from the technical idea of ​​the present disclosure.

[0015] (Embodiment) Hereinafter, a material straightening system 1 according to the present embodiment and an integrated system including the material straightening system 1 will be described with reference to FIGS.

[0016] (1) Overview As shown in Fig. 2, the material straightening system 1 according to this embodiment receives a supply of a metallic plate material 200, which is a coil material, from an uncoiler 3. The material straightening system 1 straightens the plate material 200 supplied from the uncoiler 3 so as to make it flat, and supplies the straightened plate material 200 to a press machine 4. The press machine 4 deforms the supplied plate material 200 into the shape of a die.

[0017] In the material straightening device 10 as the material straightening system 1 of the present embodiment, roll groups R1 and R2 each including a plurality of rolls 130 and 131 face each other, and the plate material 200 is passed between the facing roll groups R1 and R2 to straighten the warp of the plate material 200. At this time, in the material straightening system 1, the roll groups R1 and R2 facing each other include a plurality of rolls 130 and a plurality of rolls 131, respectively, in a staggered arrangement. In this embodiment, the roll groups R1 and R2 are arranged to face each other in the vertical direction. The roll group R1 includes a plurality of rolls 130 as shown in FIG. 3. The plurality of rolls 130 are arranged in line along the traveling direction of the plate material 200. The roll group R2 includes a plurality of rolls 131 as shown in FIG. 3. The plurality of rolls 131 are arranged in line along the traveling direction of the plate material 200. The plurality of rolls 130 and the plurality of rolls 131 are arranged in a staggered arrangement.

[0018] Here, one of the roll groups R1 and R2 may be referred to as the first straightening roll group R1. The other of the roll groups R1 and R2 may be referred to as the second straightening roll group R2. In this embodiment, the roll group arranged on the upper side of the roll groups R1 and R2 will be described as the first straightening roll group R1, and the roll group arranged on the lower side will be described as the second straightening roll group R2.

[0019] As shown in Fig. 1, the material straightening device 10 as the material straightening system 1 includes an acquisition unit 101 and an estimation unit 102. The acquisition unit 101 acquires at least one of the following parameters based on the detection result of the sensor 20 that detects information related to the parameters: the initial warpage amount, which is the warpage amount of the plate material 200 before straightening, the plate thickness of the plate material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the load applied to the first straightening roll group R1 (hereinafter, "roll load"). The estimation unit 102 estimates the warpage amount after straightening using a learning model that estimates the warpage amount after straightening, which is the warpage amount of the plate material 200 after straightening, as an objective variable, using the parameters acquired by the acquisition unit 101 as explanatory variables.

[0020] As shown in FIG. 1, the material straightening device 10 as the material straightening system 1 includes an acquisition unit 101 and a learning unit 106. The acquisition unit 101 acquires at least one of the following parameters based on the detection result of the sensor 20 that detects information related to the parameters: the initial warpage amount, which is the warpage amount of the plate material 200 before straightening, the plate thickness of the plate material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1. The acquisition unit 101 acquires the measured warpage amount, which is the warpage amount measured for the plate material 200 after straightening. The learning unit 106 learns using the parameters acquired by the acquisition unit 101 as explanatory variables and the measured warpage amount as a target variable, to generate a learning model.

[0021] The integrated system 2 of this embodiment includes a material straightening device 10 (material straightening system 1) and a sensor 20. The sensor 20 detects information related to parameters acquired by the material straightening device 10.

[0022] According to the material straightening device 10 and the integrated system 2, parameters based on the detection results of the sensor 20 are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material 200 to be straightened can be accurately estimated.

[0023] (2) Composition As shown in Fig. 1, the integrated system 2 includes a material straightening device 10 as a material straightening system 1, and a plurality of sensors 20. The plurality of sensors 20 includes a first sensor 21, a second sensor 22, a first position sensor 23, a second position sensor 24, a first pressure sensor 25, and a second pressure sensor 26. As shown in Fig. 1, the integrated system 2 further includes a third sensor 30.

[0024] The material straightening device 10, the sensors 20 and the third sensor 30 will be described below.

[0025] (2.1) Material straightening device As shown in Figs. 1 and 3, the material straightening device 10 includes a first straightening roll group R1, a second straightening roll group R2, a first roll group R3, and a second roll group R4.

[0026] The first roll group R3 guides (sends out) the plate material 200 supplied from the uncoiler 3 to the first straightening roll group R1 and the second straightening roll group R2. The first roll group R3 includes four rolls 121 to 124. The rolls 121 and 122 are arranged side by side at a predetermined first interval along the traveling direction of the plate material 200. The rolls 123 and 124 are arranged side by side at a predetermined first interval along the traveling direction of the plate material 200. The rolls 121 and 123 are arranged side by side at a predetermined second interval along a direction perpendicular to the traveling direction (here, the vertical direction). The rolls 122 and 124 are arranged side by side at a predetermined second interval along the vertical direction. Here, the predetermined second interval is equal to the thickness of the plate material 200. The rolls 121 and 122 rotate in the same direction. The rolls 123 and 124 rotate in a direction opposite to the rotation direction of the rolls 121 and 122, thereby sending out the plate material 200 to the first straightening roll group R1 and the second straightening roll group R2.

[0027] The first straightening roll group R1 includes a plurality of rolls 130, and the second straightening roll group R2 includes a plurality of rolls 131. In the first straightening roll group R1 and the second straightening roll group R2, the plurality of rolls 130 included in the first straightening roll group R1 and the plurality of rolls 131 included in the second straightening roll group R2 are vertically arranged in a staggered manner. The first straightening roll group R1 and the second straightening roll group R2 straighten the warp of the plate material 200 by passing the plate material 200 between the first straightening roll group R1 and the second straightening roll group R2.

[0028] The first straightening roll group R1 has a plurality of rolls 130 (seven in FIG. 3). The first straightening roll group R1 is provided on the lower surface of the first plate 140. The second straightening roll group R2 has a plurality of rolls 131 (seven in FIG. 3). The second straightening roll group R2 is provided on the upper surface of the second plate 141 facing the first plate 140 in the up-down direction. The first straightening roll group R1 is provided on the lower surface of the first plate 140, and the second straightening roll group R2 is provided on the upper surface of the second plate 141, so that the first straightening roll group R1 and the second straightening roll group R2 are arranged to face each other. The multiple rolls 130 rotate in the same direction, and the multiple rolls 131 rotate in the opposite direction to the rotation direction of the multiple rolls 130, so that the plate material 200 is sent to the second roll group R4. At this time, since the multiple rolls 130 of the first straightening roll group R1 and the multiple rolls 131 of the second straightening roll group R2 are arranged in a staggered pattern, the warp of the plate material 200 is straightened when it passes between the first straightening roll group R1 and the second straightening roll group R2, and the plate material 200 is sent to the second roll group R4.

[0029] The second roll group R4 guides (sends out) the plate material 200 that has passed through the first straightening roll group R1 and the second straightening roll group R2 to the press machine 4. The second roll group R4 includes four rolls 151 to 154. The rolls 151 and 152 are arranged side by side at a predetermined third interval along the traveling direction of the plate material 200. The rolls 153 and 154 are arranged side by side at a predetermined third interval along the traveling direction of the plate material 200. The rolls 151 and 153 are arranged side by side at a predetermined second interval along a direction (here, the vertical direction) perpendicular to the traveling direction. The rolls 152 and 154 are arranged side by side at a predetermined second interval along the vertical direction. Here, the predetermined second interval is equivalent to the thickness of the plate material 200. For example, the rolls 151 and 152 rotate in the same direction, and the rolls 153 and 154 rotate in the opposite direction to the rolls 151 and 152 , thereby sending the plate material 200 to the press machine 4 .

[0030] Further, as shown in FIG. 1, the material correction device 10 includes a communication unit 11, a storage unit 12, a control unit 13, and a display unit 14.

[0031] The material straightening apparatus 10 has a computer system having, for example, a processor and a memory. The processor executes a program stored in the memory, causing the computer system to function as the control unit 13. The program executed by the processor is pre-recorded in the memory of the computer system here, but may be provided by being recorded on a recording medium such as a memory card, or may be provided via an electric communication line such as the Internet.

[0032] The communication unit 11 has an interface for communicating with external devices via a network such as the Internet.

[0033] The storage unit 12 is composed of a device selected from a ROM (Read Only Memory), a RAM (Random Access Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), and the like. The storage unit 12 stores a learning model for estimating the amount of warpage corrected (amount of warpage after straightening). The learning model is, for example, a surrogate model generated by machine learning. In the learning model stored in the storage unit 12, at least one of the amount of warpage before straightening of the plate material 200 (initial amount of warpage), the thickness of the plate material 200, the insertion side position of the first straightening roll group R1, the angle (roll angle) of the first straightening roll group R1 with respect to the second straightening roll group R2, and the roll load applied to the first straightening roll group R1 is input as a parameter. In the learning model, the amount of warpage after straightening of the plate material 200 is estimated as a response variable using the input parameters. In this embodiment, the initial warpage of the sheet material 200, the thickness of the sheet material 200, the insertion side position of the first straightening roll group R1, the angle of the first straightening roll group R1 relative to the second straightening roll group R2, and the roll load applied to the first straightening roll group R1 are input to the learning model as parameters. That is, in this embodiment, five parameters are input to the learning model. Note that the learning model for estimating the straightened warpage is not limited to the surrogate model. Other learning models may be used.

[0034] Here, the roll load is a force applied to the top and bottom of the plate material 200, and depends on the material strength, plate thickness, initial warpage, the insertion side position of the first straightening roll group R1, and the roll angle. The warpage after straightening is estimated from at least one of the material strength, plate thickness, initial warpage, the insertion side position of the first straightening roll group R1, and the roll angle. In this embodiment, the warpage after straightening is estimated from five parameters, namely, the material strength, plate thickness, initial warpage, the roll position of the first straightening roll group R1, and the roll angle. Therefore, in the learning model of this embodiment, when five parameters, namely, the initial warpage of the plate material 200, the plate thickness of the plate material 200, the roll position of the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1, are input, the material strength is calculated. In the learning model, the amount of warp after straightening is estimated based on the calculated material strength, the input initial amount of warp of the plate material 200, the plate thickness of the plate material 200, the insertion side position of the first straightening roll group R1, and the roll angle.

[0035] As shown in FIG. 1, the control unit 13 includes an acquisition unit 101, an estimation unit 102, an adjustment unit 103, a notification unit 104, an output unit 105, and a learning unit 106.

[0036] The acquiring unit 101 acquires at least one of the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1, as parameters based on the detection result of the sensor 20 that detects information related to the parameters. In this embodiment, the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1 are all acquired as parameters based on the detection result of the sensor 20.

[0037] The acquiring unit 101 acquires, based on the detection results of the first sensor 21 and the second sensor 22, the amount of initial warp of the plate material 200 and the plate thickness of the plate material 200 as parameters.

[0038] The acquisition unit 101 acquires at least one of the position of the portion in the first straightening roll group R1 where the plate material 200 is inserted (insertion side position) and the roll angle as the position of the first straightening roll group R1 based on the detection results of the first position sensor 23 and the second position sensor 24. In this embodiment, the acquisition unit 101 acquires both the insertion side position and the roll angle in the first straightening roll group R1 as the position of the first straightening roll group R1 based on the detection results of the first position sensor 23 and the second position sensor 24.

[0039] The acquiring unit 101 acquires each of at least two positions in the first straightening roll group R1 along the passing direction of the sheet material 200. That is, the acquiring unit 101 acquires a position based on the detection result of the first position sensor 23 and a position based on the detection result of the second position sensor 24. Here, the position based on the detection result of the first position sensor 23 and the position based on the detection result of the second position sensor 24 are positions based on the reduction amount at both ends of the first straightening roll group R1 along the passing direction of the sheet material 200. The acquiring unit 101 acquires, as a parameter, a position obtained from a position sensor of the first position sensor 23 and the second position sensor 24 that is close to a portion where the sheet material 200 is inserted in the first straightening roll group R1. Furthermore, the acquiring unit 101 calculates (acquires) an angle (roll angle) of the first straightening roll group R1 with respect to the second straightening roll group R2 using the position based on the detection result of the first position sensor 23 and the position based on the detection result of the second position sensor 24.

[0040] The acquisition unit 101 acquires the roll load applied to the first straightening roll group R1 based on the detection results of the first pressure sensor 25 and the second pressure sensor 26. In this embodiment, the parameter as the load applied to the first straightening roll group R1 includes the detection results of both the first pressure sensor 25 and the second pressure sensor 26. That is, the acquisition unit 101 acquires, as the roll load applied to the first straightening roll group R1, each of the roll loads applied to at least two locations in the passing direction of the plate material 200 in the first straightening roll group R1.

[0041] Furthermore, the acquiring unit 101 acquires an actual amount of warpage, which is an amount of warpage measured on the straightened plate material 200. Specifically, the acquiring unit 101 acquires the actual amount of warpage based on the measurement result of the third sensor 30.

[0042] The estimation unit 102 estimates the amount of warpage of the plate material 200 after straightening using a learning model that estimates the amount of warpage of the plate material 200 after straightening, with the parameters acquired by the acquisition unit 101 as explanatory variables. Here, the estimation unit 102 estimates the amount of warpage of the plate material 200 after straightening using a learning model stored in the storage unit 12, with the initial amount of warpage of the plate material 200 acquired as parameters by the acquisition unit 101, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1 as explanatory variables.

[0043] The adjustment unit 103 adjusts the parameters acquired by the acquisition unit 101 among the position of the first straightening roll group R1 and the roll load applied to the first straightening roll group R1, using the learning model stored in the storage unit 12, so as to flatten the sheet material 200. When the position of the first straightening roll group R1 is the adjustment target, the adjustment unit 103 adjusts at least one of the insertion side position and the roll angle in the first straightening roll group R1. That is, when the position of the first straightening roll group R1 is the adjustment target, the adjustment unit 103 determines at least one of the insertion side position and the roll angle in the first straightening roll group R1. When the roll load applied to the first straightening roll group R1 is the adjustment target, the adjustment unit 103 calculates the roll load applied to the first straightening roll group R1.

[0044] For example, the adjustment unit 103 changes at least one value of the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. The adjustment unit 103 applies a learning model based on the changed value to estimate the warpage of the plate material 200. The adjustment unit 103 repeats the change and estimation of the value to determine the set values ​​of the parameters for flattening the plate material 200, that is, the initial warpage of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. The initial warpage of the plate material 200 and the plate thickness of the plate material 200 are the same as the values ​​acquired by the acquisition unit 101. The estimation of the warpage of the plate material 200 based on the changed value may be performed by the estimation unit 102.

[0045] The notification unit 104 notifies the adjusted parameters of the adjustment target, which are the parameters adjusted by the adjustment unit 103. The notification unit 104 notifies each of the set values ​​of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. Specifically, the notification unit 104 causes the display unit 14 to display each of the set values ​​of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1.

[0046] The output unit 105 outputs to the outside the parameters acquired by the acquisition unit 101, the amount of warp after correction estimated by the estimation unit 102, and the adjusted parameters of the adjustment target adjusted by the adjustment unit 103. The output unit 105 outputs to an external device via a network the parameters acquired by the acquisition unit 101, the amount of warp after correction estimated by the estimation unit 102, and the parameters of the adjustment target adjusted by the adjustment unit 103. For example, the external device can manage the parameters acquired by the acquisition unit 101, the amount of warp after correction estimated by the estimation unit 102, and the adjusted parameters of the adjustment target adjusted by the adjustment unit 103.

[0047] The learning unit 106 performs re-learning with the parameters acquired by the acquisition unit 101 as explanatory variables and the measured amount of warping acquired by the acquisition unit 101 as an objective variable, and updates the learning model. That is, the learning unit 106 performs learning with the parameters acquired by the acquisition unit 101 as explanatory variables and the measured amount of warping as an objective variable, and generates a learning model.

[0048] For example, when the error between the amount of warpage after correction estimated by the estimation unit 102 and the amount of warpage actually measured by the acquisition unit 101 is equal to or greater than a predetermined value, the control unit 13 shifts from the inference phase, in which inference is performed using the learning model, to the learning phase, in which the learning model is reconstructed, in response to a user operation. The learning unit 106 updates (generates) the learning model in the learning phase.

[0049] The display unit 14 is a thin display device such as a liquid crystal display or an organic EL (electroluminescence) display. The display unit 14 displays the adjusted parameters of the adjustment target adjusted by the adjustment unit 103. The display unit 14 displays each set value of the initial warp amount of the sheet material 200, the sheet thickness of the sheet material 200, the insertion side position in the first straightening roll group R1, the angle (roll angle) of the first straightening roll group R1 with respect to the second straightening roll group R2, and the roll load applied to the first straightening roll group R1, which are output from the notification unit 104. The user of the material straightening device 10 who sees each of the above set values ​​displayed on the display unit 14 adjusts at least one of the parameters of the position of the first straightening roll group R1 and the roll load applied to the first straightening roll group R1 based on the displayed contents. That is, the user of the material straightening device 10 controls at least one of the parameters of the position of the first straightening roll group R1 and the roll load applied to the roll group R1 based on the displayed contents.

[0050] (2.2) Sensors 1, the multiple sensors 20 included in the integrated system 2 include a first sensor 21, a second sensor 22, a first position sensor 23, a second position sensor 24, a first pressure sensor 25, and a second pressure sensor 26. The integrated system 2 further includes a third sensor 30 as shown in FIG.

[0051] The first sensor 21 and the second sensor 22 are sensors that detect the distance to a detection target (here, the plate material 200). The first sensor 21 is disposed on the opposite side of the rolls 121 and 122 from the rolls 123 and 124, and between the rolls 121 and 122 when viewed from the arrangement direction of the rolls 121 and 123 (see FIG. 3). The second sensor 22 is disposed on the opposite side of the rolls 123 and 124 from the rolls 121 and 122, and between the rolls 123 and 124 when viewed from the arrangement direction of the rolls 121 and 123 (see FIG. 3). The positions of the laser irradiation of the plate material 200 by the first sensor 21 and the second sensor 22 are on the same straight line along the thickness direction of the plate material 200, that is, along the arrangement direction of the rolls 121 and 123.

[0052] The first sensor 21 and the second sensor 22 irradiate a laser toward the plate material 200 and measure the time from the emission of the laser to the reception of the reflected wave, thereby detecting the distance to the plate material 200. The acquisition unit 101 obtains the amount of warping of the plate material 200 from the change in the distance to the plate material 200 detected by the first sensor 21, and obtains it as the initial amount of warping. The acquisition unit 101 obtains the thickness of the plate material 200 by subtracting the distance to the plate material 200 detected by the first sensor 21 and the distance to the plate material 200 detected by the second sensor 22 from the distance between the first sensor 21 and the second sensor 22. The acquisition unit 101 may obtain the amount of warping of the plate material 200 from the change in the distance to the plate material 200 detected by the second sensor 22, and obtain it as the initial amount of warping. That is, the first sensor 21 and the second sensor 22 detect information related to the initial amount of warping and the plate thickness as parameters.

[0053] The first position sensor 23 and the second position sensor 24 are disposed along the arrangement direction of the rolls 130 of the first straightening roll group R1 (the traveling direction of the sheet material 200). The first position sensor 23 is disposed closer to the portion (end) of the first straightening roll group R1 where the sheet material 200 is inserted than the second position sensor 24 (see FIG. 3). The first position sensor 23 and the second position sensor 24 have a crowning cylinder (hydraulic drive). The first position sensor 23 and the second position sensor 24 are configured to be able to adjust the distance between each of both ends of the first straightening roll group R1 along the arrangement direction of the rolls 130 of the first straightening roll group R1 (the traveling direction of the sheet material 200) and the second straightening roll group R2 by the crowning cylinder. The first position sensor 23 and the second position sensor 24 detect the amount of reduction by the crowning cylinder. That is, the first position sensor 23 and the second position sensor 24 detect information related to the position of the first straightening roll group R1 as a parameter.

[0054] The first pressure sensor 25 and the second pressure sensor 26 are respectively arranged along the arrangement direction of the rolls 130 of the first straightening roll group R1 (the traveling direction of the sheet material 200). The first pressure sensor 25 is arranged at an end of the first plate 140 closer to the portion where the sheet material 200 is inserted in the first straightening roll group R1 than the second pressure sensor 26 (see FIG. 3). The second pressure sensor 26 is arranged at an end of the first plate 140 closer to the portion where the sheet material 200 is output in the first straightening roll group R1 than the first pressure sensor 25. The first pressure sensor 25 and the second pressure sensor 26 detect the load applied to the first straightening roll group R1 at the portion where they are arranged. That is, the first pressure sensor 25 and the second pressure sensor 26 detect information related to the roll load as a parameter.

[0055] The third sensor 30 is a sensor that detects the distance to a detection object (here, the plate material 200) like the first sensor 21 and the second sensor 22. The third sensor 30 is disposed on the opposite side of the rolls 153, 154 with respect to the rolls 151, 152, and between the rolls 151 and 152 when viewed in the arrangement direction of the rolls 151 and 153 (see FIG. 3). The third sensor 30 detects the distance to the plate material 200 by irradiating the plate material 200 with a laser and measuring the time from the emission of the laser to the reception of the reflected wave. The acquisition unit 101 obtains the amount of warping of the plate material 200 from the change in the distance to the plate material 200 detected by the third sensor 30, and acquires it as the measured amount of warping. The third sensor 30 may be disposed on the opposite side of the rolls 153 and 154 from the rolls 151 and 152, and between the rolls 153 and 154 when viewed from the direction in which the rolls 151 and 153 are arranged.

[0056] (3) Operation Here, the operation of the material straightening device 10 will be described with reference to FIGS.

[0057] (3.1) Inference Phase Operations First, the operation of the material straightening apparatus 10 in the inference phase will be described with reference to FIG.

[0058] The acquisition unit 101 performs a first acquisition process (step S1). In the first acquisition process, the acquisition unit 101 acquires at least one of the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the insertion side position in the first straightening roll group R1, the angle (roll angle) of the first straightening roll group R1 relative to the second straightening roll group R2, and the roll load applied to the first straightening roll group R1 as a parameter based on the detection result of the sensor 20 that detects the information related to the parameter. Here, the acquisition unit 101 acquires the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1 as parameters based on the detection result of each sensor 20. The acquisition unit 101 acquires the initial warp amount of the sheet material 200 and the thickness of the sheet material 200 based on the detection results of the first sensor 21 and the second sensor 22. The acquisition unit 101 acquires the position (insertion side position) of the portion in the first straightening roll group R1 where the plate material 200 is inserted and the roll angle based on the detection results of the first position sensor 23 and the second position sensor 24. The acquisition unit 101 acquires the roll load acting on the first straightening roll group R1 based on the detection results of the first pressure sensor 25 and the second pressure sensor 26.

[0059] The estimation unit 102 performs estimation processing (step S2). In the estimation processing, the estimation unit 102 estimates the amount of warpage of the plate material 200 after straightening using a learning model stored in the storage unit 12, with the parameters acquired by the acquisition unit 101 as explanatory variables. Here, the estimation unit 102 estimates the amount of warpage of the plate material 200 after straightening using a learning model with the initial amount of warpage of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1 as explanatory variables.

[0060] The adjustment unit 103 performs an adjustment process (step S3). In the adjustment process, the adjustment unit 103 uses a learning model to adjust the parameters of at least one of the adjustment targets, the position of the first straightening roll group R1 and the roll load applied to the first straightening roll group R1, so that the plate material 200 is flattened. For example, the adjustment unit 103 changes at least one value of the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. The adjustment unit 103 applies the learning model based on the changed value to estimate the amount of warping of the plate material 200. The adjustment unit 103 repeats the change and estimation of the values, and determines the set values ​​of the parameters for flattening the plate material 200, that is, the initial amount of warping of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. As described above, the estimation unit 102 may estimate the amount of warpage of the plate material 200 based on the changed value.

[0061] The notification unit 104 performs notification processing (step S4). In the notification processing, the notification unit 104 notifies the adjusted parameters of the adjustment target adjusted by the adjustment unit 103. The notification unit 104 notifies each set value of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1. For example, the notification unit 104 displays each set value of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1 on the display unit 14. The user of the material straightening device 10, who has seen each of the above set values ​​displayed on the display unit 14, adjusts at least one parameter of the position of the first straightening roll group R1 and the roll load applied to the first straightening roll group R1 based on the displayed contents. For example, when adjusting the position of the first straightening roll group R1, the user controls the positions of both ends along the arrangement direction of the multiple rolls 130 of the first straightening roll group R1 (the traveling direction of the plate material 200) by using the crowning cylinders of the first position sensor 23 and the second position sensor 24. In addition, when adjusting the roll load, the user controls the roll load related to the first straightening roll group R1.

[0062] The output unit 105 performs an output process (step S5). In the output process, the output unit 105 outputs to the outside the parameters acquired by the acquisition unit 101, the post-correction warpage amount estimated by the estimation unit 102, and the post-adjustment parameters of the adjustment target adjusted by the adjustment unit 103.

[0063] (3.2) Behavior during the learning phase First, the operation of the material straightening apparatus 10 in the learning phase will be described with reference to FIG.

[0064] If the error between the amount of warping after correction estimated by the estimation unit 102 and the actual amount of warping acquired by the acquisition unit 101 is equal to or greater than a predetermined value, the control unit 13, upon operation by the user, transitions from the inference phase in which inference is made using a learning model to the learning phase in which the learning model is reconstructed.

[0065] The acquiring unit 101 performs a second acquiring process in the learning phase (step S11). In the second acquiring process, the acquiring unit 101 acquires at least one of the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the insertion side position in the first straightening roll group R1, the angle (roll angle) of the first straightening roll group R1 with respect to the second straightening roll group R2, and the roll load applied to the first straightening roll group R1 as a parameter based on the detection result of the sensor 20 that detects the information related to the parameter. Here, the acquiring unit 101 acquires each of the initial warp amount of the sheet material 200, the thickness of the sheet material 200, the insertion side position in the first straightening roll group R1, the roll angle, and the roll load applied to the first straightening roll group R1 as a parameter based on the detection result of each sensor 20. The acquiring unit 101 acquires the initial warpage amount of the sheet material 200 and the sheet thickness of the sheet material 200 based on the detection results of the first sensor 21 and the second sensor 22. The acquiring unit 101 acquires the position (insertion side position) of the portion where the sheet material 200 is inserted in the first straightening roll group R1 and the roll angle based on the detection results of the first position sensor 23 and the second position sensor 24. The acquiring unit 101 acquires the roll load applied to the first straightening roll group R1 based on the detection results of the first pressure sensor 25 and the second pressure sensor 26. Furthermore, the acquiring unit 101 acquires the actual warpage amount based on the measurement result of the third sensor 30.

[0066] The learning unit 106 performs a learning process (step S12). In the learning process, the learning unit 106 performs learning (re-learning) using the parameters acquired by the acquisition unit 101 as explanatory variables and the measured warpage amount acquired by the acquisition unit 101 as a target variable, and generates (updates) a learning model.

[0067] (4) Advantages As described above, in the material straightening system 1 according to the present embodiment, roll groups R1 and R2 each including a plurality of rolls (roll 130, roll 131) face each other, and the plate material 200 is passed between the facing roll groups R1 and R2 to straighten the warp of the plate material 200. A plurality of rolls 130 included in one roll group R1 of the facing roll groups R1 and a plurality of rolls 131 included in the other roll group R2 are arranged in a staggered pattern. The material straightening system 1 includes an acquisition unit 101 and an estimation unit 102. The acquisition unit 101 acquires information related to the parameters based on the detection result of the sensor 20 that detects at least one of the following parameters: the initial warp amount, which is the amount of warp of the plate material 200 before straightening, the thickness of the plate material 200, the position of the first straightening roll group R1, which is one roll group of the facing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1. The estimation unit 102 estimates the amount of warpage after straightening using a learning model that uses the parameters acquired by the acquisition unit 101 as explanatory variables and estimates the amount of warpage after straightening, which is the amount of warpage of the plate material 200 after straightening, as a response variable.

[0068] In addition, in the material straightening system 1 according to the present embodiment, roll groups R1 and R2 each including a plurality of rolls (roll 130, roll 131) face each other, and the plate material 200 is passed between the facing roll groups R1 and R2 to straighten the warp of the plate material 200. A plurality of rolls 130 included in one roll group R1 of the facing roll groups R1 and a plurality of rolls 131 included in the other roll group R2 are arranged in a staggered pattern. The material straightening system 1 includes an acquisition unit 101 and a learning unit 106. The acquisition unit 101 acquires information related to the parameters based on the detection result of a sensor 20 that detects at least one of the following parameters: an initial warp amount, which is the amount of warp of the plate material 200 before straightening, a plate thickness of the plate material 200, a position of a first straightening roll group R1, which is one of the facing roll groups R1 and R2, and a roll load applied to the first straightening roll group R1. The acquiring unit 101 acquires an actual warpage amount, which is an amount of warpage measured on the plate material 200 after the straightening. The learning unit 106 performs learning using the parameters acquired by the acquiring unit 101 as explanatory variables and the actual warpage amount as a target variable, and generates a learning model.

[0069] Furthermore, the integrated system 2 according to this embodiment includes the material correction system 1 and a sensor 20 that detects information related to parameters acquired by the material correction system 1.

[0070] According to the material straightening system 1 and the integrated system 2, parameters based on the detection results of the sensor 20 are used in a learning model that estimates the amount of warpage after straightening, so that the amount of warpage after straightening of the plate material 200 to be straightened can be accurately estimated.

[0071] (5) Variations The following are examples of modified examples. The modified examples described below can be applied in appropriate combination with the above-described embodiment.

[0072] (5.1) Variation 1 In the first modification, the material straightening device 10A as the material straightening system 1 controls the adjustment target based on the adjusted parameters of the adjustment target adjusted by the adjustment unit 103. The following mainly describes the differences. Note that the same reference numerals are used for the same components as in the embodiment, and the description thereof will be omitted.

[0073] As shown in Fig. 6, the integrated system 2 includes a material straightening apparatus 10A as a material straightening system 1, and a plurality of sensors 20. The plurality of sensors 20 includes a first sensor 21, a second sensor 22, a first position sensor 23, a second position sensor 24, a first pressure sensor 25, and a second pressure sensor 26. As shown in Fig. 1, the integrated system 2 further includes a third sensor 30.

[0074] The material straightening device 10A includes a first straightening roll group R1, a second straightening roll group R2, a first roll group R3, and a second roll group R4, as shown in Fig. 6. Furthermore, the material straightening device 10A includes a communication unit 11, a storage unit 12, a control unit 13A, and a display unit 14, as shown in Fig. 6.

[0075] The material straightening apparatus 10A has a computer system having, for example, a processor and a memory. The processor executes a program stored in the memory, and the computer system functions as the control unit 13A. The program executed by the processor is pre-recorded in the memory of the computer system here, but may be provided by being recorded on a recording medium such as a memory card, or may be provided via an electric communication line such as the Internet.

[0076] The control unit 13A includes an acquisition unit 101, an estimation unit 102, an adjustment unit 103, an output unit 105, a learning unit 106, and a control processing unit 107, as shown in FIG.

[0077] The control processing unit 107 controls the adjustment target based on the adjusted parameters of the adjustment target adjusted by the adjustment unit 103. For example, the control processing unit 107 adjusts the position and roll load of the first straightening roll group R1 using an actuator so that the position and roll load of the first straightening roll group R1 become the adjusted parameters adjusted by the adjustment unit 103.

[0078] (5.2) Variation 2 In the above embodiment, the estimation unit 102 is configured to estimate the amount of warpage after straightening using the learning model stored in the storage unit 12 included in the material straightening apparatus 10, but is not limited to this configuration.

[0079] The estimation unit 102 may acquire the learning model from outside and estimate the amount of warpage after correction. In this case, the output unit 105 transmits the parameters acquired by the acquisition unit 101, the amount of warpage after correction estimated by the estimation unit 102, and the measured amount of warpage acquired by the acquisition unit 101 to an external device having the learning model. The external device performs learning (re-learning) using the parameters acquired by the acquisition unit 101, the amount of warpage after correction estimated by the estimation unit 102, and the measured amount of warpage acquired by the acquisition unit 101, and generates (updates) the learning model.

[0080] (5.3) Variation 3 In the embodiment, the integrated system 2 is configured to include the first pressure sensor 25 and the second pressure sensor 26 as the sensor 20 for detecting the roll load, but is not limited to this configuration.

[0081] The number of sensors 20 (pressure sensors) that detect the roll load may be one, or three or more.

[0082] When there is one pressure sensor, the pressure sensor is provided in the center of the first plate 140 along the traveling direction of the plate material 200. The acquisition unit 101 acquires the roll pressure based on the detection result of the pressure sensor.

[0083] When there are three or more pressure sensors, the three or more pressure sensors are provided at a predetermined interval on the first plate 140 along the traveling direction of the sheet material 200. Furthermore, one pressure sensor of the three or more pressure sensors is arranged at an end of the first plate 140 closer to the portion where the sheet material 200 is inserted in the first straightening roll group R1 than the remaining pressure sensors. Furthermore, another pressure sensor of the three or more pressure sensors is arranged at an end of the first plate 140 closer to the portion where the sheet material 200 is output in the first straightening roll group R1 than the remaining pressure sensors. The acquisition unit 101 acquires the roll pressure based on the detection results of each pressure sensor.

[0084] (5.4) Variation 4 In the embodiment, the integrated system 2 is configured to include the first position sensor 23 and the second position sensor 24 as the sensor 20 for detecting the position of the first straightening roll group R1, but is not limited to this configuration.

[0085] The number of the sensor 20 (position sensor) that detects the position of the first straightening roll group R1 may be one, or three or more.

[0086] In the case where there is one position sensor, the position sensor is provided at a position closer to the portion where the sheet material 200 is inserted in the first straightening roll group R1 than the portion where the sheet material 200 is output. The acquisition unit 101 acquires the roll pressure based on the detection result of the position sensor. In this case, the angle (roll angle) of the first straightening roll group R1 with respect to the second straightening roll group R2 is not necessary as a parameter to be input to the learning model. In other words, the roll angle is not used when estimating the amount of warpage after straightening.

[0087] When there are three or more position sensors, the three or more position sensors are arranged at a predetermined interval along the arrangement direction of the multiple rolls 130 of the first straightening roll group R1 (the traveling direction of the sheet material 200). Furthermore, one of the three or more position sensors is arranged closer to the portion (end) where the sheet material 200 is inserted in the first straightening roll group R1 than the remaining position sensors. Another of the three or more position sensors is arranged closer to the portion where the sheet material 200 is output in the first straightening roll group R1 than the remaining position sensors. The acquisition unit 101 acquires at least one of the position of the portion where the sheet material 200 is inserted in the first straightening roll group R1 (insertion side position) and the roll angle based on the detection results of each position sensor.

[0088] (5.5) Variation 5 In the embodiment, the acquiring unit 101 is configured to acquire, as the position of the first straightening roll group R1, the position based on the detection result of the first position sensor 23 and the angle of the first straightening roll group R1 with respect to the second straightening roll group R2. However, the invention is not limited to this configuration.

[0089] The acquiring unit 101 may acquire a position based on the detection result of the first position sensor 23 and a position based on the detection result of the second position sensor 24 as the position of the first straightening roll group R1. That is, the acquiring unit 101 acquires each of at least two positions in the first straightening roll group R1 along the passing direction of the plate material 200. In this case, the learning model uses the position based on the detection result of the first position sensor 23 and the position based on the detection result of the second position sensor 24 to obtain the angle of the first straightening roll group R1 with respect to the second straightening roll group R2.

[0090] (5.6) Variation 6 In the embodiment, as an example of the estimation, the initial warp amount of the sheet material 200, the sheet thickness of the sheet material 200, the position of the first straightening roll group R1, and the roll load applied to the first straightening roll group R1 are each configured as parameters. However, the estimation is not limited to this configuration.

[0091] Depending on the material of the plate material 200, there may be cases where it is not necessary to use as parameters each of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the position of the first straightening roll group R1, and the roll load applied to the first straightening roll group R1. Therefore, the material straightening system 1 (material straightening device 10) estimates the warp amount after straightening using at least one of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the position of the first straightening roll group R1, and the roll load applied to the first straightening roll group R1 as a parameter.

[0092] (5.7) Variation 7 In the embodiment, the first straightening roll group R1 and the second straightening roll group R2 are arranged to face each other in the vertical direction, but are not limited to this configuration. The first straightening roll group R1 and the second straightening roll group R2 may be arranged to face each other in a direction perpendicular to the vertical direction.

[0093] (5.8) Variation 8 In the embodiment, the load is applied to the first straightening roll group R1, but the present invention is not limited to this configuration. The load may be applied to the second straightening roll group R2. In this case, the first pressure sensor 25 and the second pressure sensor 26 detect the load applied to the second straightening roll group R2.

[0094] (5.9) Variation 9 The material straightening device 10 (material straightening system 1) may store the parameters acquired by the acquisition unit 101 in the storage unit 12. For example, the material straightening device 10 stores at least one of the initial warp amount of the plate material 200, the plate thickness of the plate material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1, acquired by the acquisition unit 101, as a parameter.

[0095] The material straightening apparatus 10 may then generate a learning model based on the stored parameters.

[0096] (Other variations) The above embodiment is merely one of various embodiments of the present disclosure. The above embodiment can be modified in various ways depending on the design and the like as long as the object of the present disclosure can be achieved. In addition, the same function as the material straightening system 1 may be embodied in a material straightening method, a computer program, or a non-transitory recording medium on which a program is recorded. The material straightening method of the material straightening system 1 according to one aspect is executed by one or more processors (computer systems). The material straightening method is a material straightening method used in the material straightening system 1. In the material straightening system 1, roll groups R1 and R2 including a plurality of rolls (rolls 130 and 131) face each other, and the plate material 200 is passed between the facing roll groups R1 and R2 to straighten the warp of the plate material 200. A plurality of rolls 130 included in one roll group R1 of the facing roll groups R1 and R2 and a plurality of rolls 131 included in the other roll group R2 are arranged in a staggered manner. The material straightening method includes an acquisition step and an estimation step. The acquisition step acquires at least one of the initial warpage amount, which is the warpage amount of the plate material 200 before straightening, the plate thickness of the plate material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1, based on the detection result of the sensor 20 that detects information related to the parameter. The estimation step estimates the warpage amount after straightening using a learning model that estimates the warpage amount after straightening, which is the warpage amount of the plate material 200 after straightening, as the objective variable, using the parameters acquired in the acquisition step as explanatory variables. In addition, a material straightening method of the material straightening system 1 according to another embodiment is executed by one or more processors (computer systems). The material straightening method is a material straightening method used in the material straightening system 1. In the material straightening system 1, roll groups R1 and R2 including a plurality of rolls (rolls 130 and 131) face each other, and the plate material 200 is straightened by passing it between the opposing roll groups R1 and R2. Of the opposing roll groups R1 and R2, one roll group R1 includes a plurality of rolls 130 and the other roll group R2 includes a plurality of rolls 131, which are arranged in a staggered manner. The material straightening method includes an acquisition step and a learning step.The acquisition step acquires at least one of the initial warpage amount, which is the warpage amount of the plate material 200 before straightening, the plate thickness of the plate material 200, the position of the first straightening roll group R1, which is one of the opposing roll groups R1 and R2, and the roll load applied to the first straightening roll group R1, based on the detection result of a sensor that detects information related to the parameter. The acquisition step acquires the measured warpage amount, which is the warpage amount measured for the plate material 200 after straightening. The learning step performs learning using the parameters acquired in the acquisition step as explanatory variables and the measured warpage amount acquired in the acquisition step as a target variable to generate a learning model. The program according to one aspect is a program for causing a computer system to function as the above-mentioned material straightening system 1 or an equipment abnormality estimation method for the material straightening system 1.

[0097] The material straightening system 1 or the material straightening method of the material straightening system 1 in the present disclosure includes a computer system. The computer system has a processor and a memory as hardware. The processor executes a program recorded in the memory of the computer system, thereby realizing the function of the material straightening system 1 or the material straightening method of the material straightening system 1 in the present disclosure. The program may be pre-recorded in the memory of the computer system, or may be provided through an electric communication line. The program may be recorded in a non-transitory recording medium such as a memory card, an optical disk, or a hard disk drive that can be read by the computer system and provided. The processor of the computer system is composed of one or more electronic circuits including a semiconductor integrated circuit (IC) or a large scale integrated circuit (LSI). The integrated circuits such as IC or LSI here are called by different names depending on the degree of integration, and include integrated circuits called system LSI, VLSI (Very Large Scale Integration), or ULSI (Ultra Large Scale Integration). Furthermore, a field-programmable gate array (FPGA) that is programmed after the manufacture of an LSI, or a logic device that allows reconfiguration of the connection relationship within the LSI or reconfiguration of the circuit partition within the LSI, can also be employed as a processor. The multiple electronic circuits may be integrated into one chip, or may be distributed among multiple chips. The multiple chips may be integrated into one device, or may be distributed among multiple devices.

[0098] In addition, it is not essential for the material straightening system 1 that multiple functions are integrated into one housing, and the components of the material straightening system 1 may be distributed across multiple housings. Furthermore, at least some of the functions of the material straightening system 1 may be realized by cloud (cloud computing) or the like.

[0099] (summary) As described above, in the material straightening system (1) of the first embodiment, roll groups (R1, R2) including a plurality of rolls (roll 130, roll 131) face each other, and the plate material (200) is passed between the facing roll groups (R1, R2) to straighten the warp of the plate material (200). Among the facing roll groups (R1, R2), a plurality of rolls (130) included in one roll group (R1) and a plurality of rolls (131) included in the other roll group (R2) are arranged in a staggered pattern. The material straightening system (1) includes an acquisition unit (101) and an estimation unit (102). The acquisition unit (101) acquires at least one of the initial warpage, which is the amount of warpage of the plate material (200) before straightening, the thickness of the plate material (200), the position of the first straightening roll group (R1), which is one of the opposing roll groups (R1, R2), and the roll load applied to the first straightening roll group (R1), as parameters based on the detection results of the sensor (20) that detects information related to the parameters. The estimation unit (102) estimates the amount of warpage after straightening using a learning model that uses the parameters acquired by the acquisition unit (101) as explanatory variables and estimates the amount of warpage after straightening, which is the amount of warpage of the plate material (200) after straightening, as an objective variable.

[0100] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy.

[0101] The material straightening system (1) of the second aspect is the first or second aspect, and further includes a learning unit (106). The acquisition unit (101) further acquires an actual warpage amount, which is an amount of warpage measured on the straightened plate material (200). The learning unit (106) performs re-learning using the parameters acquired by the acquisition unit (101) as explanatory variables and the actual warpage amount as a target variable, thereby updating the learning model.

[0102] According to this embodiment, it is possible to improve the accuracy of estimating the amount of warpage of the plate material (200) to be straightened after the straightening.

[0103] The material straightening system (1) of the third aspect is any one of the first to third aspects, and further includes an adjustment unit (103). The acquisition unit (101) acquires at least one of the position of the first straightening roll group (R1) and the roll load applied to the first straightening roll group (R1) as a parameter. The adjustment unit (103) adjusts the parameter to be adjusted using a learning model, with at least one of the position of the first straightening roll group (R1) and the roll load applied to the first straightening roll group (R1) acquired by the acquisition unit (101) as an adjustment target so that the plate material (200) is flattened.

[0104] According to this embodiment, the parameters to be adjusted are adjusted based on the estimation result, so that the plate material (200) can be made more flat.

[0105] The material correction system (1) of the fourth aspect is the third aspect, further comprising a notification unit (104). The notification unit (104) notifies the post-adjustment parameters of the adjustment target adjusted by the adjustment unit (103).

[0106] According to this embodiment, the user who receives the notification can further flatten the plate material (200) by adjusting the parameters to be adjusted.

[0107] The material straightening system (1) of the fifth aspect is the third or fourth aspect, further comprising a control processing unit (107). The control processing unit (107) controls the adjustment target based on the adjusted parameters of the adjustment target adjusted by the adjustment unit (103).

[0108] According to this embodiment, the plate material (200) can be made more flat by controlling the object to be adjusted.

[0109] The material straightening system (1) of the sixth aspect is in any one of the third to fifth aspects, and further includes an output unit (105). The output unit (105) outputs to the outside the parameters acquired by the acquisition unit (101), the amount of warpage after straightening estimated by the estimation unit (102), and the adjusted parameters of the adjustment target adjusted by the adjustment unit (103).

[0110] According to this embodiment, it is possible to externally manage the parameters acquired by the acquisition unit (101), the post-correction warpage amount estimated by the estimation unit (102), and the post-adjustment parameters of the adjustment target adjusted by the adjustment unit (103). Furthermore, it is also possible to externally generate a learning model based on this information.

[0111] In the seventh aspect of the material straightening system (1), in any of the first to sixth aspects, the acquisition section (101) acquires, as the positions of the first straightening roll group (R1), the positions of at least two locations in the first straightening roll group (R1) along the passing direction of the plate material (200).

[0112] According to this embodiment, the position of the first straightening roll group (R1) can be specified more accurately, so that the accuracy of estimating the amount of warpage after straightening can be improved.

[0113] In the eighth aspect of the material straightening system (1), in any of the first to seventh aspects, the acquisition section (101) acquires, as the roll loads applied to the first straightening roll group (R1), each of the roll loads applied to at least two locations in the passing direction of the plate material (200) in the first straightening roll group (R1).

[0114] According to this embodiment, the roll load acting on the first straightening roll group (R1) can be more accurately specified, so that the accuracy of estimating the amount of warpage after straightening can be improved.

[0115] In the material straightening system (1) of the ninth aspect, in any one of the first to eighth aspects, the estimation unit (102) acquires the learning model from outside.

[0116] According to this embodiment, the information used for the estimation can be distributed.

[0117] In the material straightening system (1) of the tenth aspect, roll groups (R1, R2) including a plurality of rolls (roll 130, roll 131) face each other, and the plate material (200) is passed between the facing roll groups (R1, R2) to straighten the warp of the plate material (200). Among the facing roll groups (R1, R2), a plurality of rolls (130) included in one roll group (R1) and a plurality of rolls (131) included in the other roll group (R2) are arranged in a staggered pattern. The material straightening system (1) includes an acquisition unit (101) and a learning unit (106). The acquisition unit (101) acquires at least one of the parameters, an initial warpage, which is the amount of warpage of the plate material (200) before straightening, the thickness of the plate material (200), the position of the first straightening roll group (R1), which is one of the opposing roll groups (R1, R2), and the roll load applied to the first straightening roll group (R1), based on the detection result of the sensor (20) that detects information related to the parameters. The acquisition unit (101) acquires an actual warpage, which is the amount of warpage measured on the plate material (200) after straightening, and the amount of warpage. The learning unit (106) performs learning using the parameters acquired by the acquisition unit (101) as explanatory variables and the actual warpage as a target variable, to generate a learning model.

[0118] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy.

[0119] The material straightening method of the eleventh aspect is used in a material straightening system (1). In the material straightening system (1), roll groups (R1, R2) including a plurality of rolls (roll 130, roll 131) face each other, and the plate material (200) is passed between the facing roll groups (R1, R2) to straighten the warp of the plate material (200). Among the facing roll groups (R1, R2), a plurality of rolls (130) included in one roll group (R1) and a plurality of rolls (131) included in the other roll group (R2) are arranged in a staggered manner. The material straightening method includes an acquisition step and an estimation step. The acquisition step acquires at least one of the initial warpage, which is the amount of warpage of the plate material (200) before straightening, the thickness of the plate material (200), the position of the first straightening roll group (R1), which is one of the opposing roll groups (R1, R2), and the roll load applied to the first straightening roll group (R1), as a parameter based on the detection result of the sensor (20) that detects information related to the parameter. The estimation step estimates the amount of warpage after straightening using a learning model that estimates the amount of warpage after straightening, which is the amount of warpage of the plate material (200) after straightening, as a target variable, using the parameters acquired in the acquisition step as explanatory variables.

[0120] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy.

[0121] The material straightening method of the twelfth aspect is used in a material straightening system (1). In the material straightening system (1), roll groups (R1, R2) including a plurality of rolls (roll 130, roll 131) face each other, and the plate material (200) is passed between the facing roll groups (R1, R2) to straighten the warp of the plate material (200). Among the facing roll groups (R1, R2), a plurality of rolls (130) included in one roll group (R1) and a plurality of rolls 131 included in the other roll group (R2) are arranged in a staggered pattern. The material straightening method includes an acquisition step and a learning step. The acquisition step acquires at least one of the initial warpage, which is the amount of warpage of the plate material (200) before straightening, the thickness of the plate material (200), the position of the first straightening roll group (R1), which is one of the opposing roll groups (R1, R2), and the roll load applied to the first straightening roll group (R1), based on the detection result of the sensor (20) that detects information related to the parameter. The acquisition step acquires the measured warpage, which is the amount of warpage measured for the plate material (200) after straightening. The learning step performs learning using the parameters acquired in the acquisition step as explanatory variables and the measured warpage acquired in the acquisition step as a target variable, to generate a learning model.

[0122] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy.

[0123] A program according to a thirteenth aspect is a program for causing a computer system to execute the material straightening method according to the eleventh or twelfth aspect.

[0124] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy.

[0125] The integrated system (2) of the fourteenth aspect includes the material straightening system (1) of any one of the first to tenth aspects, and a sensor (20) that detects information related to parameters acquired by the material straightening system (1).

[0126] According to this embodiment, parameters based on the detection results of the sensor (20) are used in a learning model that estimates the amount of warping after straightening, so that the amount of warping after straightening of the plate material (200) to be straightened can be estimated with high accuracy. [Explanation of symbols]

[0127] 1. Material straightening system 2. Integrated System 20 Sensors 101 Acquisition Department 102 Estimation part 103 Adjustment section 104 Notification Department 105 Output section 106 Learning Department 107 Control processing section 130 rolls 131 rolls 200 Board material R1 First straightening roll group (roll group) R2 Second straightening roll group (roll group)

Claims

1. A material straightening system in which roll groups each including a plurality of rolls are opposed to each other, and a plurality of rolls included in one of the opposed roll groups and a plurality of rolls included in the other roll group are arranged in a staggered manner, and a plate material is passed between the opposed roll groups to straighten the warp of the plate material, an acquisition unit that acquires information related to at least one of an initial warpage, which is an amount of warpage of the plate material before straightening, a plate thickness of the plate material, a position of a first straightening roll group, which is one of the opposing roll groups, and a roll load applied to the first straightening roll group, based on a detection result of a sensor that detects information related to the parameters; and an estimation unit that estimates the amount of warpage after straightening by using a learning model that estimates the amount of warpage after straightening, which is the amount of warpage of the plate material after straightening, as an explanatory variable, using the parameters acquired by the acquisition unit as explanatory variables, Material straightening system.

2. Further comprising a learning section; The acquisition unit further acquires an actual warpage amount, which is an amount of warpage measured on the plate material after the straightening, the learning unit performs re-learning using the parameters acquired by the acquisition unit as explanatory variables and the measured warpage as a target variable, thereby updating the learning model.

10. The material straightening system of claim 1.

3. Further comprising an adjustment unit, The acquisition unit acquires at least one of a position of the first straightening roll group and a roll load applied to the first straightening roll group as the parameter, the adjustment unit adjusts the parameters of the position of the first straightening roll group and the roll load applied to the first straightening roll group acquired by the acquisition unit as adjustment targets using the learning model so that the plate material is flattened.

10. The material straightening system of claim 1.

4. a notification unit that notifies the adjustment target of a post-adjustment parameter, the post-adjustment parameter being the parameter adjusted by the adjustment unit; 4. A material straightening system according to claim 3.

5. A control processing unit that controls the adjustment target based on an adjusted parameter of the adjustment target adjusted by the adjustment unit.

4. A material straightening system according to claim 3.

6. an output unit configured to output to an outside the parameter acquired by the acquisition unit, the amount of warpage after correction estimated by the estimation unit, and the adjusted parameter of the adjustment target adjusted by the adjustment unit, 4. A material straightening system according to claim 3.

7. The acquisition unit acquires, as the positions of the first straightening roll group, positions of at least two locations in the first straightening roll group along a passing direction of the plate material.

10. The material straightening system of claim 1.

8. The acquisition unit acquires, as the roll loads applied to the first straightening roll group, roll loads applied to at least two locations along a passing direction of the plate material in the first straightening roll group.

10. The material straightening system of claim 1.

9. The estimation unit acquires the learning model from an external source.

10. The material straightening system of claim 1.

10. A material straightening system in which roll groups each including a plurality of rolls are opposed to each other, and a plurality of rolls included in one of the opposed roll groups and a plurality of rolls included in the other roll group are arranged in a staggered manner, and a plate material is passed between the opposed roll groups to straighten the warp of the plate material, an acquisition unit that acquires information related to at least one of an initial warpage, which is the amount of warpage of the plate material before straightening, a thickness of the plate material, a position of a first straightening roll group, which is one of the opposing roll groups, and a roll load applied to the first straightening roll group, based on a detection result of a sensor that detects the information, and acquires an actual warpage, which is the amount of warpage measured on the plate material after straightening; a learning unit that performs learning using the parameters acquired by the acquisition unit as explanatory variables and the measured warpage as a target variable to generate a learning model, Material straightening system.

11. A material straightening method used in a material straightening system in which roll groups each including a plurality of rolls are opposed to each other, and a plurality of rolls included in one roll group and a plurality of rolls included in the other roll group are arranged in a staggered manner, and a plate material is passed between the opposed roll groups to straighten the warp of the plate material, an acquisition step of acquiring information related to at least one of an initial warpage amount, which is an amount of warpage of the plate material before straightening, a plate thickness of the plate material, a position of a first straightening roll group, which is one of the opposing roll groups, and a roll load applied to the first straightening roll group, based on a detection result of a sensor that detects information related to the parameter; and an estimation step of estimating the amount of warpage after straightening using a learning model that estimates the amount of warpage after straightening, which is the amount of warpage of the plate material after straightening, as an explanatory variable using the parameters acquired in the acquisition step as explanatory variables, Material straightening method.

12. A material straightening method used in a material straightening system in which roll groups each including a plurality of rolls are opposed to each other, and a plurality of rolls included in one roll group and a plurality of rolls included in the other roll group are arranged in a staggered manner, and a plate material is passed between the opposed roll groups to straighten the warp of the plate material, an acquisition step of acquiring information related to at least one of an initial warpage amount, which is the warpage amount of the plate material before straightening, a plate thickness of the plate material, a position of a first straightening roll group, which is one of the opposing roll groups, and a roll load applied to the first straightening roll group, based on a detection result of a sensor that detects information related to the parameters, and acquiring an actual warpage amount, which is the warpage amount measured on the plate material after straightening; A learning step of performing learning using the parameters acquired in the acquisition step as explanatory variables and the measured warpage amount acquired in the acquisition step as a response variable to generate a learning model. Material straightening method.

13. A program for causing a computer system to execute the material straightening method according to claim 11 or 12.

14. A material straightening system according to any one of claims 1 to 10; A sensor for detecting information related to the parameter obtained by the material straightening system. Integrated system.

Citation Information

Patent Citations

  • Estimation method for deformation state of material to be corrected and control method for roll push-in amount of roller leveller

    JP2020175443A

Cited By

  • Artificial intelligence-based planarization process control system

    JP7892865B1