Position correction information processing device

Abstract

To propose a position correction information processing device that can be used in various formats of press work machines and enables highly accurate processing according to a state of a workpiece or a mold.SOLUTION: A position correction information processing device 20 controls a position correction device 30 that corrects a position of a mold component 51 according to a state of at least one of a workpiece M or a mold 50. The position correction information processing device 20 includes: an input information acquisition part 21 that acquires input information 21 including at least one of workpiece state information 11a indicating a state of the workpiece M and mold state information 12a indicating a state of the mold 50; and a correction information output part 22 that outputs correction information 22a for controlling the position correction device 30 on the basis of the input information 21a.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a position correction information processing device that corrects the positions of die components used in a press machine. [Background technology]

[0002] A press die has been disclosed that includes a punch die having a fixed punch and a movable punch, and a die die having a bending blade and a pad, in which the punch die and the die die come close to each other to bend a steel plate into a hat cross-sectional shape, and when the movable punch moves from the close position to the separated position, tension is applied to the vertical wall portion, and the relative distance between the movable punch and the pad is fixed at a predetermined distance at the separated position (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-161171 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the press die in Patent Document 1 is a technology used only for press forming hat-shaped cross-section parts. Also, the measurement targets only the relative distance between the movable punch and the pad, and does not take into account the state of the workpiece or the die.

[0005] An object of the present invention is to provide a position correction information processing device that can be used in various types of press machines and that is capable of performing high-precision processing based on the state of the workpiece or mold. [Means for solving the problem]

[0006] The position correction information processing device according to the present invention comprises: A position correction information processing device that controls a position correction device that corrects the position of a die component based on at least one state of a workpiece and a die, an input information acquisition unit that acquires input information including at least one of workpiece information indicating a state of the workpiece and die information indicating a state of the die; a correction information output unit that outputs correction information for controlling the position correction device based on the input information; Equipped with. [Effects of the Invention]

[0007] The position correction information processing device according to the present invention can be used in various types of press machines, and enables high-precision processing based on the state of the workpiece or die. [Brief explanation of the drawings]

[0008] [Figure 1] 1 shows an example of a system configuration of a position correction system 1 according to a first embodiment. [Figure 2] 1 shows an example of a mold 50 in which the position correction device 30 of the first embodiment is installed. [Figure 3] 1 shows an example of various types of information used in the position correction system 1 of the first embodiment. [Figure 4] 1 shows an example of a position correction device 30 according to the first embodiment. [Figure 5] 5 shows a view taken along arrows VV in FIG. 4. [Figure 6] 6 shows a view taken along the line VI-VI in FIG. 5. [Figure 7] 1 shows an example of an exploded perspective view of a position correction device 30 according to a first embodiment before assembly. [Figure 8] 10 shows an example of a method for acquiring correction information 22a in the position correction information processing device 20 of the first embodiment. [Figure 9] 3 shows an example of a flowchart of a position correction method according to the first embodiment. [Figure 10] 1 shows an example of a first example of an electric pressing machine 100. [Figure 11] 1 shows an example of a slide mechanism of the electric press machine 100 of the first example. [Figure 12]1 shows an example of a horizontal cross section including a slide 1111 of the electric pressing machine 100 of the first example. [Figure 13] 1 shows an example of a system configuration of an electric press machine 100 according to a first example. [Figure 14] 1 shows an example of a flowchart of press working by the electric press machine 100 of the first example. [Figure 15] 1 shows an example of the relationship between the corrected position and the thickness of the processed material in the electric press machine 100 of the first example. [Figure 16] An example of a second example of an electric press machine 200 is shown. [Figure 17] 10 shows an example of the vicinity of a die 50 of an electric press machine 200 of a second example. [Figure 18] 10 shows an example of the system configuration of a machine learning device 60 of a position correction system 1 according to a second embodiment. [Figure 19] 10 shows an example of a neural network model used in the machine learning device 60 of the second embodiment. [Figure 20] 10 shows an example of a flowchart of a machine learning method performed by a machine learning device 60 according to the second embodiment. [Figure 21] 10 shows an example of a system configuration of a position correction information processing device 20 according to a second embodiment. [Figure 22] 10 shows an example of a flowchart of a correction information inference method according to a second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Fig. 1 shows an example of the system configuration of a position correction system 1 according to the first embodiment. Fig. 2 shows an example of a mold 50 in which a position correction device 30 according to the first embodiment is installed. Fig. 3 shows examples of various information used in the position correction system 1 according to the first embodiment. Fig. 3(a) shows an example of input information 21a, Fig. 3(b) shows an example of workpiece characteristic information 40a and mold characteristic information 40b, and Fig. 3(c) shows an example of correction information 22a.

[0010] The mold 50 shown in Fig. 1 has a structure for a progressive crushing mold 50 that performs multiple processes while feeding a material. The mold 50 may be used for any process, such as shearing, drawing, or bending. It may also be a mold 50 used for processes such as punching or cutting a multilayer film.

[0011] The position correction system 1 includes an information input device 10 installed on the mold 50, a position correction information processing device 20 that calculates correction information 22a based on input information 21a input from the information input device 10, a position correction device 30 that corrects the position of the punch 51 of the mold 50 based on the correction information 22a calculated by the position correction information processing device 20, and a database device 40 that stores at least one characteristic of the workpiece M or the mold 50.

[0012] The mold 50 has an upper mold 50a and a lower mold 50b, and is equipped with at least the position correction device 30 of the position correction system 1. The information input device 10 and the position correction information processing device 20 may be installed in the mold 50, or may be installed in other positions or components.

[0013] The information input device 10 may be a sensor or the like that measures the state of at least one of the workpiece M or the die 50. The measurement values are processed and transmitted as input information 21a to the position correction information processing device 20. The information input device 10 has a workpiece information input unit 11 and a die information input unit 12. The workpiece information input unit 11 inputs workpiece state information 11a, and the die information input unit 12 inputs die state information 12a.

[0014] The workpiece condition information 11a may be at least one of a pre-processing workpiece thickness indicating the thickness of the workpiece M before processing or a post-processing workpiece thickness indicating the thickness of the workpiece M after processing. For example, the pre-processing workpiece thickness may be measured by sandwiching the pre-processing workpiece M between a pressing member installed on the upper die 50a and the workpiece M and using a thickness sensor or the like as the workpiece information input unit 11. The post-processing workpiece thickness may be measured by sandwiching the post-processing workpiece M between a pressing member installed on the upper die 50a and the workpiece M and using a thickness sensor or the like as the workpiece information input unit 11.

[0015] The die state information 12a may be at least one of the load applied to at least a portion of the die 50 or the temperature of at least a portion of the die 50. For example, the die load may be the load applied to the punch 51 during processing, which may be measured using a load sensor or the like serving as the die information input unit 12 installed on the upper die 50a. Furthermore, the die temperature may be the temperature of the die 50 during processing, which may be measured using a temperature sensor or the like serving as the die information input unit 12 that measures the temperature of the lower die 50b below or vertically below the punch 51.

[0016] The position correction information processing device 20 has a processing unit 20a that outputs correction information 22a based on input information 21a acquired from the information input device 10, and a correction storage unit 20b that stores an arithmetic program, an arithmetic expression, an arithmetic table, or the like used in the processing of the processing unit 20a. The position correction information processing device 20 may be installed inside the mold 50 or at a location remote from the mold 50. When installed at a location remote from the mold 50, the position correction information processing device 20 is connected to the information input device 10 and the position correction device 30 by wired or wireless communication, and transmits and receives data such as the input information 21a or the correction information 22a.

[0017] The processing unit 20a includes an input information acquisition unit 21 that acquires input information 21a and a correction information output unit 22 that outputs correction information 22a. The processing unit 20a may be configured with one or more arithmetic processing devices (CPU, MPU, GPU, DSP, etc.). The processing unit 20a may be, for example, a circuit board centered around a CPU (Central Processing Unit), or a dedicated circuit board. The processing unit 20a is supplied with power from a commercial power source via a power line (not shown).

[0018] The correction storage unit 20b stores various data and programs and may be configured, for example, with a volatile memory (DRAM, SRAM, etc.) that functions as a main memory and a non-volatile memory (ROM, flash memory, etc.). The correction storage unit 20b may be included in the processing unit 20a. Alternatively, it may be installed separately from the position correction information processing device 20.

[0019] The input information acquisition unit 21 acquires input information 21a from the information input device 10. The input information 21a may be at least one of workpiece state information 11a or die state information 12a input from at least one of the workpiece information input unit 11 or die information input unit 12 of the information input device 10. That is, the input information 21a may be at least one of the workpiece thickness before processing, the workpiece thickness after processing, the die load, or the die temperature.

[0020] The correction information output unit 22 outputs correction information 22a acquired by the calculation program, calculation formula, calculation table, or the like stored in the correction storage unit 20b, based on the input information 21a acquired by the input information acquisition unit 21. The correction information 22a may be a correction command value for correcting the position of a die part such as the punch 51.

[0021] The database device 40 stores at least one of workpiece property information 40a indicating the properties of the workpiece M and die property information 40b indicating the properties of the die 50.

[0022] The workpiece characteristic information 40a includes at least one of the following: the name of the material of the workpiece M; strength characteristics indicating the shrinkage of the workpiece M under load; and thermal characteristics indicating the shrinkage of the workpiece M under temperature. The material of the workpiece M may be steel, special steel, non-ferrous metal, non-metal, etc. For example, the steel may be hot-rolled steel plate or cold-rolled steel plate; the special steel may be special-purpose steel such as alloy steel or stainless steel; and the non-ferrous metal may be aluminum alloy, copper alloy, titanium, magnesium, etc. Non-metals may be resin or ceramic. Materials with multiple names are preferably assigned a unified name. For example, the database device 40 may store dimensions of the workpiece M that change depending on load or temperature. Furthermore, the database device 40 preferably stores the allowable upper and lower limits for the thickness of the finished product after processing the workpiece M.

[0023] The mold characteristic information 40b includes at least one of the name of the material of the mold 50, strength characteristics indicating the shrinkage of the mold 50 in response to load, and thermal characteristics indicating the shrinkage of the mold 50 in response to temperature. The material of the mold 50 may be JIS steel grade, martensitic stainless steel, non-ferrous metal, composite material, cemented carbide, etc. It is preferable to determine a unified name for materials with multiple names. For example, the database device 40 may store dimensions of the mold 50 that change depending on load or temperature.

[0024] Fig. 4 shows an example of the position correction device 30 of the first embodiment. Fig. 5 shows a view taken along arrows VV in Fig. 4. Fig. 6 shows a view taken along arrows VI-VI in Fig. 5.

[0025] The position correction device 30 of the position correction system 1 includes a nut lifting sleeve 31 attached to the punch 51 of the upper die 50a, a storage body 32 that stores the nut lifting sleeve 31 in a form that protrudes toward the punch 51, a nut lifting plate 33 that engages with the nut lifting sleeve 31 and the storage body 32 and rotates to move the nut lifting sleeve 31 slightly in its axial direction, and a worm 34 that rotates the nut lifting plate 33.

[0026] The nut lifting plate 33 is provided with worm wheel teeth 33a that mesh with a worm 34 provided in the housing 32. A spiral sliding groove 31a that progresses at a slight angle is formed in the central portion of the outer circumferential surface of the nut lifting sleeve 31, and a guide engagement portion 33b that slides and engages with the spiral sliding groove 31a of the nut lifting sleeve 31 is provided on the inner circumferential surface of the nut lifting plate 33.

[0027] The housing 32 is formed by a housing member 32a and a ring member 32b. The housing member 32a rotatably supports the worm 34 and has a stepped hole 32c drilled in its center. An annular space is formed by the step of the hole 32c and the ring member 32b fixed to the upper surface of the housing member 32a. The nut lift plate 33 is fitted into the spiral sliding groove 31a of the nut lift sleeve 31 and is housed in the annular space so as to be rotatable but constrained in the axial direction. The ring member 32b supports the outer peripheral surface of the nut lift sleeve 31 so as to be slidable in the axial direction, thereby housing the nut lift sleeve 31.

[0028] When the worm 34 rotates, the nut lifting plate 33 is rotated via the worm wheel tooth piece 33a that meshes with the worm 34, and the guide engagement part 33b rotates. That is, the guide engagement part 33b rotates along the spiral sliding groove 31a provided in the nut lifting sleeve 31, and the nut lifting sleeve 31 moves slightly in its axial direction, i.e., in the vertical direction.

[0029] A preset number of pulse voltages, for example, are applied to the motor 41 that rotates the worm 34. This causes the motor 41 to rotate a predetermined amount, causing the nut lifting sleeve 31 to move slightly in its axial direction via the rotation of the nut lifting plate 33. This movement of the nut lifting sleeve 31 moves the punch 51 up and down via the housing 32.

[0030] Fig. 7 shows an example of an exploded perspective view of the position correction device 30 according to the first embodiment before assembly. Note that in Fig. 7, the inclination and the like are exaggerated to make the explanation of the nut elevating sleeve 31 and the nut elevating plate 33 easier to understand.

[0031] The storage body 32 in which the nut lifting sleeve 31 and the nut lifting plate 33 shown in Figure 7 are stored in a combined form is composed of a roughly circular storage member 32a with a stepped hole 32c drilled in the center, and a ring member 32b that is fixed to the upper end surface of the storage member 32a after the nut lifting sleeve 31 and the nut lifting plate 33 are stored in the combined state in the hole 32c, as shown in Figures 4 to 6.

[0032] A rotatably supported worm 34 is provided inside the storage member 32a, and the worm 34 is engaged with worm wheel teeth 33a formed on a portion of the outer surface of the nut lifting plate 33, so that the nut lifting plate 33 is rotated by a motor 41 attached to the outside of the storage member 32a.

[0033] The nut-lifting sleeve 31 has an opening 31b in the center and a recess 31c around the opening 31b on its upper surface. The entire sleeve is cylindrical, with a spiral sliding groove 31a formed on its outer periphery. The sliding groove 31a divides the sleeve into an upper annular portion 31g and a lower annular portion 31h. The upper annular portion 31g is provided with a notch 31d into which a guide engaging portion 33b of the nut-lifting plate 33 is fitted, as will be described later. In the illustrated embodiment, since the nut-lifting plate 33 has two guide engaging portions 33b, two notches 31d are provided. Ends 31e and 31f are shown at the left and right ends of the notch 31d.

[0034] The nut lifting plate 33 has an opening 33c in the center and is formed in an annular shape, with worm wheel teeth 33a provided on part of the outer periphery of the annular portion 33d. In addition, two guide engagement portions 33b are provided on the inner circumferential surface of the annular portion 33d, as shown in the figure.

[0035] The guide engagement portion 33b is formed so as to be able to rotate within the sliding groove 31a while snugly engaging with the spiral sliding groove 31a in the nut elevating sleeve 31. The guide engagement portions 33b are provided on the inner peripheral surface of the annular portion 33d at an inclination angle θ corresponding to the inclined surface of the sliding groove 31a. The guide engagement portion 33b has two upper and lower planes and a vertical plane connecting these planes.

[0036] The cross-sectional shape of the guide engagement portion 33b corresponds to the cross-sectional shape of the sliding groove 31a in the above-mentioned nut elevating sleeve 31. This configuration prevents undesired rattle between the nut elevating plate 33 and the nut elevating sleeve 31 when the nut elevating plate 33 rotates within the sliding groove 31a in the nut elevating sleeve 31, and also makes it possible to fully guarantee the mechanical strength of the guide engagement portion 33b.

[0037] When engaging the nut lifting plate 33 with the nut lifting sleeve 31, the guide engagement portion 33b of the nut lifting plate 33 is aligned with the notch 31d of the nut lifting sleeve 31, and the nut lifting plate 33 is pressed against the lower annular portion 31h of the nut lifting sleeve 31, and then rotated along the sliding groove 31a of the nut lifting sleeve 31. By engaging the two in this way, the housing 32 is formed.

[0038] 6, the nut lifting sleeve 31 has two notches 31d, and a sliding groove 31a is formed between the two notches 31d. The guide engagement portion 33b of the nut lifting plate 33 engages with the sliding groove 31a. In response to the rotation of the motor 41, the two guide engagement portions 33b of the nut lifting plate 33 rotate along the two sliding grooves 31a of the nut lifting sleeve 31, respectively.

[0039] Since the sliding groove 31a in the nut lifting sleeve 31 is formed to progress in a spiral shape as described above, the nut lifting sleeve 31 moves slightly upward or downward relative to the storage member 32a in response to the rotation of the nut lifting plate 33. Note that the pin 36 shown in Fig. 6 prohibits the nut lifting sleeve 31 from rotating about the axis relative to the storage member 32a, but allows it to move upward or downward about the axis.

[0040] FIG. 8 shows an example of a method for acquiring the correction information 22a of the position correction information processing device 20 of the first embodiment.

[0041] The position correction information processing device 20 of the first embodiment obtains the correction control value by calculation. The position correction information processing device 20 of the example shown in Fig. 8 calculates the first correction value by an equation and calculates the second correction value by a table.

[0042] The first correction value is obtained by comparing the post-processing workpiece thickness with the upper and lower allowable limits of the product thickness. The post-processing workpiece thickness may be a measurement value obtained by a sensor or the like that measures the post-processing workpiece thickness. The upper and lower allowable limits of the product thickness may be stored in advance in the correction storage unit 20b or the database device 40. For example, if the post-processing workpiece thickness is greater than the upper limit of the product thickness, the first correction value may be the value obtained by subtracting the upper limit of the allowable product thickness from the post-processing workpiece thickness. If the upper limit of the allowable product thickness is subtracted from the post-processing workpiece thickness, the position correction information processing device 20 operates the position correction device 30 to move the tip position of the punch 51 in a direction closer to the workpiece M. Furthermore, if the post-processing workpiece thickness is smaller than the lower limit of the allowable product thickness, the first correction value may be the value obtained by subtracting the post-processing workpiece thickness from the lower limit of the allowable product thickness. When the workpiece thickness after processing is subtracted from the allowable lower limit of the product thickness, the position correction information processing device 20 simply activates the position correction device 30 and moves the tip position of the punch 51 in a direction away from the workpiece M.

[0043] The second correction value is acquired based on the temperature of the mold 50. The temperature of the mold 50 may be a measurement value of a sensor or the like that measures the temperature of the mold 50. The dimensional change of the workpiece M based on the temperature of the mold 50 may be stored in advance in the correction storage unit 20b or the database device 40. For example, if the temperature of the mold 50 increases, the second correction value may be increased, and if the temperature of the mold 50 decreases, the second correction value may be decreased.

[0044] The position correction device 30 is included in the position correction system 1, and is operated based on the correction information 22a output by the position correction information processing device 20 to correct the tip position of the punch 51 of the die 50. The position correction device 30 is operated by a positionable actuator such as a correction motor 39.

[0045] FIG. 9 shows an example of a flowchart of the position correction method according to the first embodiment.

[0046] First, in step 1, the position correction information processing device 20 acquires at least one of the workpiece state information 11a or the die state information 12a (ST1). At least one of the workpiece state information 11a or the die state information 12a is measured by the information input device 10. At least one of the workpiece state information 11a or the die state information 12a is transmitted to the position correction information processing device 20 as input information 21a. In this embodiment, the workpiece state information 11a or the die state information 12a may be at least one of the thickness of the workpiece M before processing, the temperature of the die 50 during processing, the load of the punch 51 during processing, or the thickness of the workpiece M after processing. Alternatively, other parameters may be used.

[0047] Subsequently, in step 2, the position correction information processing device 20 acquires correction information 22a based on the input information 21a acquired from the information input device 10. For example, the position correction information processing device 20 of the first embodiment acquires the correction information 22a as shown in FIG.

[0048] Subsequently, in step 3, the position correction information processing device 20 controls the position correction device 30 based on the correction information 22a. The correction information 22a is transmitted to the position correction device 30 as a correction command value.

[0049] As described above, the position correction information processing device 20 of the first embodiment can be used for various types of press working, and enables high-precision working based on the state of the workpiece M or the die 50.

[0050] Next, an example of an electric press machine 100 that uses a die 50 including the position correction device 30 of this embodiment will be described.

[0051] Fig. 10 shows an example of a first example of an electric press machine 100. Fig. 11 shows an example of a slide mechanism of the first example of an electric press machine 100. Fig. 12 shows an example of a horizontal cross section including a slide 111 of the first example of an electric press machine 100. Note that the support pillars 102 and crown 103 are omitted from Fig. 11.

[0052] The electric press machine 100 includes a bed 101, a support 102, a crown 103, a scale post 104, a slide 111, a motor 120 as a drive unit, a ball screw 130 as a power transmission unit, and a position detection unit 140.

[0053] The bed 101 is a base member for placing the electric press machine 100 on the ground. The support pillars 102 are columns supported by the bed 101 and extend upward. In this embodiment, there are four support pillars 102, which are installed near the four corners of the bed 101. The crown 103 is placed on the support pillars 102 and mounts the motor 120 thereon. The bed 101, the support pillars 102, and the crown 103 form the frame of the electric press machine 100. Note that the number of support pillars 102 is not limited to four, and it is sufficient that there are at least two or more support pillars that can support the crown 103. Furthermore, the support pillars are not limited to being columnar, and may be plate-shaped.

[0054] The slide 111 is movably attached to the support 102. In this embodiment, the four corners of the slide 111 are movably installed on the support 102. The slide 111 has a slight gap or the like formed between it and the support 102, giving it a tiltable structure.

[0055] The motor 120 is mounted on the crown 103 and drives a ball screw 130, which serves as a power transmission unit. As shown in FIG. 11 , the ball screw 130 has a screw shaft 130a and a connecting portion 130b. The screw shaft 130a passes through the crown 103 and is connected to the output shaft of the motor 120. The connecting portion 130b is a ball joint or the like. Therefore, the ball screw 130 mounts the slide 111 so that the slide 111 can rotate in any direction, and transmits the driving force generated by the motor 120 to the slide 111. As a result, the motor 120 drives the slide 111.

[0056] In the first example, there are four motors 120, each corresponding to the four corners of the crown 103 and the slide 111. The four motors 120 and the four ball screws 130 operate independently. The number of motors 120 is not limited to four, but at least two or more may be used.

[0057] The position detection unit 140 is preferably a linear scale or the like that reads the scale post 104 and measures the height at which the slide 111 is positioned relative to the bed 101. In this embodiment, there are four of them, corresponding to the four corners of the slide 111. Note that at least two or more position detection units 140 are required.

[0058] The scale post 104 is attached vertically to the bed 101 at one end and the crown 103 at the other end. In this embodiment, the scale post 104 is attached to the outside of the slide 111 near the four corners.

[0059] The mold 50 has an upper mold 50a and a lower mold 50b. The upper mold 50a is installed on a slide 111 and moves together with the slide 111. The lower mold 50b is installed on a bed 101. The upper mold 50a and the lower mold 50b are arranged opposite each other, and a material is processed by being sandwiched between the upper mold 50a and the lower mold 50b. A punch 51 is installed on the upper mold 50a via a position correction device 30.

[0060] 11, the ball screw 130 of the electric press machine 100 of the first example is designated as a first ball screw 131 at the top right of the page, and counterclockwise therefrom as a second ball screw 132, a third ball screw 133, and a fourth ball screw 134. The motor 120 and the position detection unit 140 correspond to the ball screw 130 in the same manner, and the first to fourth numbers are assigned to them.

[0061] FIG. 13 shows an example of the system configuration of the electric pressing machine 100 of the first example.

[0062] The electric press machine 100 has an operation panel 190 operated by a user, a control unit 170 that drives and controls the first motor 121 to the fourth motor 124 for the first axis to the fourth axis in response to commands from the operation panel 190, and a memory unit 180 that pre-stores information regarding the drive energy for each stage to be supplied to each of the first motor 121 to the fourth motor 124.

[0063] Also, corresponding to each axis, there are provided a first servo amplifier 161 to a fourth servo amplifier 164 that receive signals from the control unit 170 and drive and control the first motor 121 to the fourth motor 124, a first encoder 151 to a fourth encoder 154 that detect the rotation speed of the first motor 121 to the fourth motor 124, and a first position detection unit 141 to a fourth position detection unit 144 that detect the position of each axis.

[0064] The control unit 170 has a command unit 170a that sends command values to the first servo amplifier 161 to the fourth servo amplifier 164 corresponding to each axis, and a calculation unit 170b that calculates the command values from the detection values of the first position detection unit 141 to the fourth position detection unit 144. The operation panel 190, the control unit 170, or the memory unit 180 may be installed separately from the electric press machine 100 using a personal computer, a mobile terminal, or the like.

[0065] Next, a description will be given of the position control of the electric press machine 100 of the first example. Note that the description will be given here for a case where the electric press machine 100 has four axes.

[0066] The electric press machine 100 of the first example automatically and repeatedly performs the operation of pressing a workpiece during the actual press working period in which a molded product is actually produced. The slide 111 can be set to a horizontal position or a predetermined inclination with high precision at each stage of the press working operation during the actual press working period. The electric press machine 100 of this embodiment preferably has a teaching period prior to the actual press working period.

[0067] FIG. 14 shows an example of a flowchart of press working by the electric press machine 100 of the first example.

[0068] First, in step 11, the user sets parameters (ST11). The parameters may be the slide initial position, slide processing position, slide speed, slide load, and the position of the punch 51 of the die 50 of the electric press machine 100.

[0069] Next, in step 12, the slide 111 is lowered at a preset speed (ST12). The slide 111 is in an initial position and is initially operated by position control. First, the operation panel 190 is operated, and a speed command is sent from the command unit 170a to the servo amplifier 160, which drives the motor 120. Note that the initial position of the slide 111 is not limited to the upper limit position, and may be another position.

[0070] Next, in step 13, it is determined whether or not the slide 111 has reached the processing position (ST13). If the slide 111 has not reached the processing position in step 13, the process returns to step 12. If the slide 111 has reached the processing position in step 13, the slide 111 is stopped in step 14 (ST14). Note that the processing position of the slide 111 is set in advance, so it is sufficient to determine whether or not the slide 111 has reached the initial position based on the detection value of the position detection unit 140, etc. The processing position of the slide 111 is not limited to the lower limit position, and may be another position.

[0071] Next, in step 15, at least one state of the workpiece M or the die 50 is measured (ST15). At least one state of the workpiece M or the die 50 may be measured by the information input device 10. The state of the workpiece M may be workpiece state information 11a including at least one of the workpiece thickness before processing or the workpiece thickness after processing. Furthermore, the state of the die 50 may be die state information 12a including at least one of the die load or the die temperature.

[0072] Next, in step 16, the slide 111 is raised (ST16). Subsequently, in step 17, it is determined whether or not the slide 111 has reached its initial position (ST17). Since the initial position of the slide 111 is set in advance, it is sufficient to determine whether or not the slide 111 has reached its initial position based on the detection value of the position detection unit 140, etc.

[0073] If it is determined in step 17 that the slide 111 has not reached the initial position, the process returns to step 16. If it is determined in step 17 that the slide 111 has reached the initial position, the process stops the ascent of the slide 111 in step 18 (ST18).

[0074] Next, in step 19, it is determined whether or not to end the processing by the electric press machine 100 (ST19). The end of processing may be set in advance in the parameter setting in step 1. For example, the determination may be made based on the number of shots, the processing time, etc.

[0075] If it is determined in step 19 that the processing by the electric press machine 100 is not to be terminated, it is determined in step 20 whether or not a predetermined number of shots for performing correction has been reached (ST20). If it is determined in step 19 that the processing by the electric press machine 100 is to be terminated, the electric press machine 100 terminates its operation.

[0076] If it is determined in step 20 that the number of shots for performing correction has not been reached, the process returns to step 12. If it is determined in step 20 that the number of shots for performing correction has been reached, the position correction information processing device 20 starts correction control in step 21 (ST21). The correction control can be performed by executing the flowchart shown in FIG.

[0077] FIG. 15 shows an example of the relationship between the corrected position of the electric pressing machine 100 of the first example and the thickness of the workpiece after processing.

[0078] In the example shown in Figure 15, the upper allowable limit for the post-machining material thickness is 3.09 mm, and the lower allowable limit is 3.078 mm. When the production volume reaches 400 pieces, the post-machining material thickness exceeds the upper allowable limit of 3.09 mm, so the position of the punch 51 is corrected from the original 36.11 mm to 36.115 mm. As a result, the post-machining material thickness returns to between the upper allowable limit of 3.09 mm and the lower allowable limit of 3.078 mm. Furthermore, when the production volume reaches 600 pieces, the post-machining material thickness falls below the lower allowable limit of 3.078 mm, so the position of the punch 51 is corrected from the original 36.115 mm to 36.11 mm.

[0079] In this way, the electric pressing machine 100 of the first example makes it possible to perform high-precision processing based on the state of the workpiece M or the die 50.

[0080] 16 shows an example of a second example of an electric press machine 200. The position correction system 1 of this embodiment is also applicable to the double-slide type electric press machine 200 shown in the second example.

[0081] In the figure, 201 is a bed, 202 is a support pillar, 203 is a crown, 204 is a scale pillar, 210 is an inner slide as the first slide, 220 is an inner motor as the first side drive source, 230 is an inner ball screw as the first side feed member, 240 is an inner position detection unit as the first side position detection member, 260 is an outer slide as the second slide, 270 is an outer motor as the second side drive source, 280 is an outer ball screw as the second second side feed member, and 290 is an outer position detection unit as the second side position detection member.

[0082] The bed 201 is a member that serves as a base for placing the electric press machine 200 on the ground. The support pillars 202 are pillars that extend upward from the bed 201. In the second example, there are four support pillars 202, which are installed at the four corners of the bed 201. The crown 203 is placed on the support pillars 202 and carries the inner motor 220 and the outer motor 270. The bed 201, the support pillars 202, and the crown 203 form a frame body of the electric press machine 200. Note that the number of support pillars 202 is not limited to four, and it is sufficient that there are at least two or more pillars that can support the crown 203. Furthermore, the support pillars are not limited to being columnar, and may be plate-shaped.

[0083] The inner slide 210 has a platform portion 210a movably attached to the support column 202 and a protrusion 210b extending downward from the platform portion 210a. In this embodiment, the four corners of the platform portion 210a are slidably attached to the support column 202, and the protrusion 210b is installed so as to extend downward from the center of the platform portion 210a. Note that multiple protrusions 210b may extend from the platform portion 210a.

[0084] The inner motor 220 is placed on the crown 203 and drives the inner ball screw 230. As shown in FIG. 16, the inner ball screw 230 has a screw shaft 230a and a nut portion 230b. The screw shaft 230a passes through the crown 203 and is connected to the output shaft of the inner motor 220. The nut portion 230b is attached to the inner slide 210 and contains circulating steel balls (not shown).

[0085] The electric press machine 200 of the second example has four inner motors 220 and four inner ball screws 230, each corresponding to the four corners of the crown 203 and the inner slide 210. The four inner motors 220 and four inner ball screws 230 operate independently. The number of inner motors 220 and four inner ball screws 230 is not limited to four, but may be at least two.

[0086] The inner position detection unit 240 is preferably a linear scale or the like that reads the scale post 204 and measures the height at which the inner slide 210 is positioned relative to the bed 201. In the second example, there are four of them, corresponding to the four corners of the inner slide 210. Note that at least two or more inner position detection units 240 are sufficient.

[0087] The outer slide 260 has a platform portion 260a movably attached to the support 202 below the inner slide 110 and a hole portion 260b through which the convex portion 210b of the inner slide 210 passes movably in the up and down direction of the platform portion 260a. In a first example, the four corners of the platform portion 260a are slidably installed on the support 202, and the hole portion 260b is provided in the center of the platform portion 260a so that the convex portion 210b of the inner slide 210 passes slidably.

[0088] The outer motor 270 is placed on the crown 203 and drives the outer ball screw 280. The outer ball screw 280 has a screw shaft 280a and a nut portion 280b. The screw shaft 280a passes through the crown 203 and the inner slide 210 and is connected to the output shaft of the outer motor 270. The nut portion 280b is attached to the outer slide 260 and contains a circulating steel ball (not shown).

[0089] In the second example, there are four outer motors 270 and four outer ball screws 280, each corresponding to the four corners of the crown 203 and the outer slide 260. The four outer motors 270 and four outer ball screws 280 operate independently. The number of outer motors 270 and four outer ball screws 280 is not limited to four, but may be at least two.

[0090] The outer position detection unit 290 is preferably a linear scale or the like that reads the scale post 204 and measures the height at which the outer slide 260 is positioned relative to the bed 201. In the second example, there are four of them, corresponding to the four corners of the outer slide 260. Note that at least two outer position detection units 290 are sufficient.

[0091] The scale post 204 is attached vertically to the bed 201 at one end and the crown 203 at the other end. In the second example, it is attached to the four outer corners of the inner slide 210 and the outer slide 260. The inner position detector 240 and the outer position detector 290 share the scale post 204. Therefore, the same number of scale posts 204, inner position detectors 240, and outer position detectors 290 are provided.

[0092] The electric press machine 200 of the second example automatically and repeatedly performs the operation of pressing the workpiece M, and is designed so that the inner slide 210 and the outer slide 260 can be kept horizontal with high precision at each stage of each press operation during the actual press processing period.

[0093] That is, during the teaching processing period prior to the actual press processing period, at each stage during the progress of each press processing shot, (i) the measurement results of the inner position detection unit 240 are taken in and the drive energy to be supplied to each of the four inner motors 220 that drive the inner slide 210 is adjusted and determined so that the inner slide 210 can be kept horizontal, and information regarding the drive energy to be supplied to each of the inner motors 220 for each stage is stored in a storage device, and (ii) the measurement results of the outer position detection unit 290 are taken in and the drive energy to be supplied to each of the four outer motors 270 that drive the outer slide 260 is adjusted and determined so that the outer slide 260 can be kept horizontal, and information regarding the drive energy to be supplied to each of the outer motors 270 for each stage is stored in a storage device.

[0094] Next, at each stage during the progress of each shot of press processing during the actual processing period, (i) driving energy is supplied to each of the inner motors 220 that drive the inner slide 210 based on the stored information, and (ii) driving energy is supplied to each of the outer motors 270 that drive the outer slide 260 based on the stored information.

[0095] Because the electric pressing machine 200 of the second example is controlled in this manner, the inner slide 210 and the outer slide 260 are kept horizontal with high precision at each stage of each pressing operation. As a result, the clearance between the slide holes at the four corners of the inner slide 210 and the support columns 202 can be set to 0.10 mm to 0.25 mm.

[0096] FIG. 17 shows an example of the vicinity of the die 50 of the electric press machine 200 of the second example.

[0097] The mold 50 used in the electric press machine 200 of the second example has an upper mold 50a and a lower mold 50b. The upper mold 50a is installed on the inner slide 210 or the outer slide 260 and moves together with the inner slide 210 or the outer slide 260. The lower mold 50b is installed on the bed 201. The upper mold 50a and the lower mold 50b are arranged opposite each other, and a material is processed by being sandwiched between the upper mold 50a and the lower mold 50b.

[0098] An inner punch 51a is installed on the upper die 50a installed on the inner slide 210 via an inner position correction device 30a. An outer punch 51b is installed on the upper die 50a installed on the outer slide 260 via an outer position correction device 30b.

[0099] Therefore, the position of the inner punch 51a can be corrected by controlling the inner position correcting device 30a, and the position of the outer punch 51b can be corrected by controlling the outer position correcting device 30b.

[0100] In this way, the electric pressing machine 100 of the second example makes it possible to perform high-precision processing based on the state of the workpiece M or the die 50.

[0101] Next, a position correction system 1 according to a second embodiment will be described.

[0102] 18 shows an example of the system configuration of the machine learning device 60 of the position correction system 1 of the second embodiment. In the position correction system 1 of the second embodiment, the information input device 10, the position correction device 30, and the database device 40 are similar to those of the position correction system 1 of the first embodiment, and therefore a description thereof will be omitted.

[0103] The position correction system 1 may be a system that actually uses data learned by this machine learning device 60 in the future, or may be a test device that simulates the same structure as the system that will actually be used.

[0104] The machine learning device 60 includes a learning data acquisition unit 61, a learning data storage unit 62, a machine learning unit 63, and a trained model storage unit 64. The machine learning device 60 is configured, for example, by a computer or the like. In this case, the learning data acquisition unit 61 is configured by a communication interface or an input / output device or the like, the machine learning unit 63 is configured by a processor or the like, and the learning data storage unit 62 and the trained model storage unit 64 are configured by storage or the like.

[0105] The learning data acquisition unit 61 is an interface unit connected to various external devices via a network 90 or the like, and acquires learning data including at least input data. The external devices are, for example, the information input device 10, an input unit provided in a simulated test device, or the like, and an operator terminal 80 used by an operator.

[0106] The learning data storage unit 62 is a database that stores one or more sets of learning data acquired by the learning data acquisition unit 61. The specific configuration of the database that constitutes the learning data storage unit 62 may be designed as appropriate.

[0107] The machine learning unit 63 performs machine learning using the learning data stored in the learning data storage unit 62. That is, the machine learning unit 63 inputs multiple sets of learning data to the learning model 64a, and causes the learning model 64a to learn the correlation between the input information 21a and the correction information 22a contained in the learning data, thereby generating a trained learning model 64a. In the second embodiment, a case will be described in which a neural network is employed as a specific method of supervised learning by the machine learning unit 63.

[0108] The trained model storage unit 64 is a database that stores the trained learning model 64a generated by the machine learning unit 63. The trained learning model 64a stored in the trained model storage unit 64 is provided to the real system via any communication network, recording medium, etc. Note that although the training data storage unit 62 and the trained model storage unit 64 are shown as separate storage units in FIG. 18, they may be configured as a single storage unit.

[0109] The learning data includes at least one of workpiece state information 11a and die state information 12a as input information 21a. When "supervised learning" is adopted as machine learning, the learning data further includes control command values for moving die components such as the punch 51 as correction information 22a associated with the input information 21a. In supervised learning, the correction information 22a is referred to as, for example, teacher data or correct answer labels.

[0110] Therefore, the learning data for the second embodiment is configured by associating input information 21a including at least one of workpiece condition information 11a or die condition information 12a with correction information 22a including control command values for moving die parts such as punch 51.

[0111] Here, the correlation between the input information 21a and the correction information 22a contained in the learning data will be described.

[0112] The control command values for moving die components such as the punch 51 are largely determined by at least one of the workpiece characteristic information 40a, which indicates the material name of the workpiece M and its characteristics, and the die characteristic information 40b, which indicates the material name of the die 50 and its characteristics. However, since the dimensions of the workpiece M or the die 50 change depending on at least one of the states of the workpiece M or the die 50, the control command values for moving the die components change from moment to moment. Therefore, by measuring at least one of the states of the workpiece M or the die 50 as input data for learning, it becomes possible to infer the control command values for moving the die components.

[0113] When acquiring the above-mentioned learning data, the learning data acquisition unit 61 uses as input data at least one of the material name and its characteristics of the workpiece M or the material name and its characteristics of the die 50 stored in the database device 40, and at least one of the states of the workpiece M or the die 50. When the worker acquires the dimensions of the workpiece M after processing and inputs them into the worker terminal 80, the learning data acquisition unit 61 acquires, as output data (teaching data), from the worker terminal 80. The control command values for moving the die components obtained based on the dimensions of the workpiece M after processing input into the worker terminal 80. The learning data acquisition unit 61 then configures one learning data set by associating the input data with the output data set, and stores the learning data set in the learning data storage unit 62.

[0114] FIG. 19 shows an example of a neural network model used in the machine learning device 60 according to the second embodiment.

[0115] The learning model 64a is configured as a neural network model shown in Fig. 14. The neural network model is configured from l neurons (x1 to x1) in the input layer, m neurons (y11 to y1m) in the first hidden layer, n neurons (y21 to y2n) in the second hidden layer, and o neurons (z1 to zo) in the output layer.

[0116] Each neuron in the input layer is associated with a respective piece of input data included in the training data. Each neuron in the output layer is associated with a respective piece of output data included in the training data. Note that the input data may be subjected to predetermined pre-processing before being input to the input layer, and the output data may be subjected to predetermined post-processing after being output from the output layer.

[0117] The first and second hidden layers are also called hidden layers, and the neural network may have multiple hidden layers in addition to the first and second hidden layers, or may have only the first hidden layer as a hidden layer. Furthermore, synapses connecting the neurons of each layer are established between the input layer and the first hidden layer, between the first hidden layer and the second hidden layer, and between the second hidden layer and the output layer, and each synapse is assigned a weight wi (i is a natural number).

[0118] A neural network model uses training data to input input data contained in the training data into the input layer, and compares the output data output from the output layer as the inference result with the output data (teacher data) contained in the training data, thereby learning the correlation between the input data and the output data.

[0119] Specifically, each neuron in the input layer receives input data included in the training data, and the value of each neuron in the output layer is calculated by performing a process for all neurons other than the input layer, in which the value of the neuron on the input side connected to the neuron in question is calculated as the sum of a series of multiplication values of the value of the neuron on the input side connected to the neuron in question and the weight wi associated with the synapse connecting the neuron on the output side and the neuron on the input side.

[0120] Then, the values (z1 to zo) output to each neuron in the output layer as the inference results are compared with the values (t1 to to) of the teacher data corresponding to each output data included in the learning data to determine the error, and a process (back propagation) is performed to adjust the weight wi associated with each synapse so that the error becomes small.

[0121] When a predetermined learning termination condition is met, such as repeating the above series of steps a predetermined number of times or the above error becoming smaller than an allowable value, the machine learning is terminated and a trained neural network model (all weights wi associated with each synapse) is generated.

[0122] FIG. 20 is a flowchart showing an example of a machine learning method performed by the machine learning device 60 according to this embodiment.

[0123] First, in step 21, the learning data acquisition unit 61 prepares a desired number of pieces of learning data as a preliminary preparation for starting machine learning, and stores the prepared learning data in the learning data storage unit 62 (ST21). The number of pieces of learning data to be prepared here may be set in consideration of the inference accuracy required for the ultimately obtained learning model 64a.

[0124] Several methods can be used to prepare learning data. For example, when inferring control command values for moving die components such as punch 51 in a specific electric press machine 100, 200 or test device, the control command values for moving the die components are acquired using the learning data acquisition unit 61, and an operator uses the operator terminal 8 to input the results in association with these measurement values, thereby preparing input data and output data that constitute the learning data. Then, by repeating this process, it is possible to prepare multiple sets of learning data.

[0125] Next, in step 22, the machine learning unit 63 prepares a pre-learning model 64a to start machine learning (ST22). The pre-learning model 64a prepared here is configured with the neural network model exemplified in FIG. 19, and the weights of each synapse are set to initial values. Each neuron in the input layer is associated with at least one of workpiece condition information 11a or die condition information 12a, and at least one of workpiece characteristic information 40a or die characteristic information 40b, as input data included in the learning data. Each neuron in the output layer is associated with a control command value for moving a die component.

[0126] Next, in step 23, the machine learning unit 63 randomly acquires, for example, one piece of learning data from the plurality of sets of learning data stored in the learning data storage unit 62 (ST23).

[0127] Next, in step 24, the machine learning unit 63 inputs input data included in one piece of learning data to the input layer of the prepared learning model 64a before learning (or during learning) (ST24). As a result, output data is output as an inference result from the output layer of the learning model 64a, and this output data was generated by the learning model 64a before learning (or during learning). Therefore, in the state before learning (or during learning), the output data output as an inference result represents information different from the output data (teacher data) included in the learning data.

[0128] Next, in step 25, the machine learning unit 63 compares the output data (teacher data) included in the one learning data acquired in step 22 with the output data output from the output layer as an inference result in step 23, and adjusts the weight of each synapse, thereby performing machine learning (ST25). In this way, the machine learning unit 63 causes the learning model 64a to learn the correlation between the input data and the output data.

[0129] Next, in step 26, the machine learning unit 63 determines whether or not it is necessary to continue machine learning based on, for example, the error between the output data and the teacher data and the remaining number of unlearned learning data stored in the learning data storage unit 62 (ST26).

[0130] In step 26, if the machine learning unit 63 determines to continue machine learning (No in step 26), the process returns to step 23 and performs steps 23 to 25 on the learning model 64a being trained multiple times using untrained training data.On the other hand, in step 26, if the machine learning unit 63 determines to end machine learning (Yes in step 26), the process proceeds to step 27.

[0131] Then, in step 27, the machine learning unit 63 stores the trained learning model 64a generated by adjusting the weights associated with each synapse in the trained model storage unit 43 (ST27), thereby completing the series of machine learning methods shown in Fig. 20. In the machine learning method, step 21 corresponds to a learning data storage step, steps 22 to 26 correspond to a machine learning step, and step 27 corresponds to a trained model storage step.

[0132] As described above, the machine learning device 60 and machine learning method according to the second embodiment can provide a learning model 64a that can infer control command values for moving die components from at least one of workpiece state information 11a or die state information 12a, and at least one of workpiece characteristic information 40a or die characteristic information 40b, as input data included in the learning data. Furthermore, the machine learning device 60 and machine learning method according to the second embodiment enable high-precision machining based on the state of the workpiece M or die 50.

[0133] FIG. 21 shows an example of the system configuration of a position correction information processing device 20 according to the second embodiment.

[0134] The position correction information processing device 20 of the second embodiment includes a processing unit 20a and a trained model storage unit 20c. The processing unit 20a includes an input information acquisition unit 21 and a correction information output unit 22. The position correction information processing device 20 is configured, for example, by a computer. In this case, the input information acquisition unit 21 is configured by a communication interface or an input / output device, the correction information output unit 22 is configured by a processor, and the trained model storage unit 20c is configured by storage. The position correction information processing device 20 may be incorporated into the die 50 or the electric press machine 100, 200, or may be incorporated into a higher-level management device of the electric press machine 100, 200 (for example, an equipment controller, an equipment management system that manages multiple pieces of equipment, etc.).

[0135] The input information acquisition unit 21 is an interface unit connected to the information input device 10, the database device 40, and the machine learning device 60, and acquires input information 21a (at least one of the workpiece condition information 11a or the mold condition information 12a, and the workpiece characteristic information 40a or the mold characteristic information 40b).

[0136] The correction information output unit 22 inputs the input information 21a acquired by the input information acquisition unit 21 into the learning model 64a and performs inference processing to infer correction information 22a of the control command value for moving the mold part. The inference processing uses the machine learning device 60 and the learned learning model 64a that has undergone supervised learning using a machine learning method.

[0137] The correction information output unit 22 not only performs inference processing using the learning model 64a, but also includes a preprocessing function of adjusting the input information 21a acquired by the input information acquisition unit 21 into a desired format and inputting the adjusted information into the learning model 64a as preprocessing of the inference processing, and a postprocessing function of applying a predetermined logical formula or calculation formula to the value of the output data output from the learning model 64a to ultimately infer a control command value for moving the mold component as postprocessing of the inference processing. Note that the inference results of the correction information output unit 22 are preferably stored in the trained model storage unit 20c or another storage device (not shown), and past inference results can be used as learning data for online learning or re-learning to further improve the inference accuracy of the learning model 64a, for example.

[0138] The trained model storage unit 20c is a database that stores trained learning models 64a used in the inference process of the correction information output unit 22. The number of learning models 64a stored in the trained model storage unit 20c is not limited to one. For example, multiple learning models 64a with different numbers of input data or different machine learning methods may be stored and selectively available.

[0139] FIG. 22 shows an example of a flowchart of the correction information inference method according to the second embodiment.

[0140] First, in step 31, the input information acquisition unit 21 acquires at least one of the workpiece condition information 11a or the die condition information 12a, and at least one of the workpiece characteristic information 40a or the die characteristic information 40b (ST31).

[0141] Next, in step 32, the correction information output unit 22 performs preprocessing on the input data and inputs the preprocessed data to the input layer of the learning model 64a, and acquires the output data output from the output layer of the learning model 64a (ST32).

[0142] Next, in step 33, the correction information output unit 22 infers a control command value for moving the die part (ST33).

[0143] As described above, the machine learning device 60 and the machine learning method according to the second embodiment enable highly accurate machining based on the state of the workpiece M or the mold 50.

[0144] The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and scope of the present invention, all of which are included in the technical concept of the present invention.

[0145] In the above embodiment, a case has been described in which a neural network is employed as a specific method of machine learning by the machine learning unit 63, but any other machine learning method may be employed by the machine learning unit 63. Examples of other machine learning methods include tree-type methods such as decision trees and regression trees, ensemble learning such as bagging and boosting, neural network-type methods (including deep learning) such as recurrent neural networks and convolutional neural networks, clustering-type methods such as hierarchical clustering, non-hierarchical clustering, k-nearest neighbors, and k-means, multivariate analyses such as principal component analysis, factor analysis, and logistic regression, and support vector machines.

[0146] The present invention can also be provided in the form of a program (machine learning program) for causing a general-purpose computer to execute each step of the machine learning method according to the above embodiment. Also, the present invention can also be provided in the form of a program (position correction information inference program) for causing a general-purpose computer to execute each step of the position correction information inference method according to the above embodiment.

[0147] The present invention can be provided not only in the form of the position correction information processing device 20 (position correction information inference method or position correction information inference program) according to the above embodiment, but also in the form of an inference device (inference method or inference program) used to infer the correction information 22a. In this case, the inference device (inference method or inference program) can include a memory and a processor, and the processor executes a series of processes. The series of processes includes an input information acquisition process (input information acquisition step) for acquiring input information 21a (at least one of workpiece state information 11a or die state information 12a, and at least one of workpiece characteristic information 40a or die characteristic information 40b), and a correction information output process (correction information output step) for inferring correction information 22a of a control command value for moving a die component once the input information 21a is acquired in the input information acquisition process.

[0148] By providing it in the form of an inference device (inference method or inference program), it can be more easily applied to various devices than when implementing the position correction information processing device 20. It is naturally understandable to those skilled in the art that when the inference device (inference method or inference program) infers the control command value for moving the mold part, it may apply the inference method implemented by the correction information output unit 22 of the position correction information processing device 20 using the machine learning device 60 and the trained learning model 64a generated by the machine learning method according to the above embodiment.

[0149] Although the position correction system 1 has been described above based on several embodiments, the present invention is not limited to these embodiments and various combinations and modifications are possible. [Explanation of symbols]

[0150] 1...Position correction system, M...Workpiece material 10...information input device, 11...workpiece information input unit, 11a...workpiece state information, 12...die information input unit, 12a...die state information, 20...position correction information processing device, 20a...processing unit, 20b...correction storage unit, 20c...trained model storage unit, 21...input information acquisition unit, 21a...input information, 22...correction information output unit, 22a...correction information, 30…Position correction device 40...Database device, 40a...Workpiece material characteristic information, 40b...Mold characteristic information, 50... die, 51... punch (die part), 51a... inner punch (die part), 51b... outer punch (die part), 60... machine learning device, 61... learning data acquisition unit, 62... learning data storage unit, 63... machine learning unit, 64... model storage unit, 64a... learning model

Claims

1. A position correction information processing device that controls a position correction device that corrects the position of a die part based on at least one state of a workpiece or a die, an input information acquisition unit that acquires input information including at least one of workpiece state information indicating a state of the workpiece or die state information indicating a state of the die; a correction information output unit that outputs correction information for controlling the position correction device based on the input information; Equipped with Position correction information processing device.

2. The workpiece state information is A workpiece thickness before processing, which indicates the thickness of the workpiece before processing; a post-processing workpiece thickness indicating the thickness of the workpiece after processing; At least one of Contains The position correction information processing device according to claim 1 .

3. The mold state information is a mold load indicative of a load on at least a portion of the mold; a mold temperature indicative of the temperature of at least a portion of the mold; At least one of Contains The position correction information processing device according to claim 1 .

4. The input information includes at least one of workpiece characteristic information indicating characteristics of the workpiece and die characteristic information indicating characteristics of the die. The position correction information processing device according to claim 1 .

5. The workpiece characteristic information is a strength characteristic indicating shrinkage of the workpiece against a load; a thermal characteristic indicating the shrinkage of the workpiece with respect to temperature; Contains at least one of The position correction information processing device according to claim 4 .

6. The mold characteristic information is a strength characteristic indicating shrinkage against a load of the mold; a thermal profile indicating shrinkage versus temperature of the mold; Contains at least one of The position correction information processing device according to claim 4 .

7. The correction information is a correction command value for controlling the position correction device; Contains The position correction information processing device according to claim 1 .

8. The correction information output unit outputs the correction information by inputting the input information into a learning model that has undergone machine learning to determine a correlation between at least one of the workpiece state information or the die state information and the correction information. The position correction information processing device according to claim 1 .

9. A position correction system that corrects a position of a die part based on at least one state of a workpiece or a die, an information input device for measuring input information including at least one of workpiece state information indicating the state of the workpiece and die state information indicating the state of the die; a position correction information processing device including an input information acquisition unit that acquires the input information, and a correction information output unit that outputs correction information for correcting the positions of the mold parts based on the input information; a position correction device that corrects the positions of the mold parts based on the correction information; Equipped with Position correction system.

10. A position correction system that corrects a position of a die part based on at least one state of a workpiece or a die, an information input device for measuring input information including at least one of workpiece state information indicating the state of the workpiece and die state information indicating the state of the die; a machine learning device including: a learning data storage unit that stores a plurality of sets of learning data each composed of the input information and correction information for correcting the position of the mold component; a machine learning unit that inputs the plurality of sets of learning data into a learning model, thereby causing the learning model to learn a correlation between the input information and the correction information; and a trained model storage unit that stores the learning model that has learned the correlation by the machine learning unit; a position correction information processing device having a correction information output unit that inputs the input information to the learning model and outputs the correction information for the input information; a position correction device that corrects the positions of the mold parts based on the correction information; Equipped with Position correction system.

11. An inference device comprising a memory and a processor, The processor: an input information acquisition process for acquiring input information including at least one of workpiece state information indicating the state of the workpiece or die state information indicating the state of the die; an inference process for inferring correction information for correcting the positions of the mold parts based on the input information acquired by the input information acquisition process; To execute Reasoning device.

12. a learning data storage unit that stores multiple sets of learning data, each set consisting of input information including at least one of workpiece state information indicating the state of the workpiece or die state information indicating the state of the die, and correction information for correcting the positions of the die components; a machine learning unit that inputs a plurality of sets of the learning data into a learning model, thereby causing the learning model to learn a correlation between the input information and the correction information; a learned model storage unit that stores the learned model in which the correlation is learned by the machine learning unit; Equipped with Machine learning device.

13. A position correction method for correcting the position of a mold part, comprising: an input information acquisition step of acquiring input information including at least one of workpiece state information indicating the state of the workpiece or die state information indicating the state of the die; a correction information output step of outputting correction information for correcting the positions of the mold parts based on the input information acquired by the input information acquisition step; Equipped with Position correction method.

14. An inference method executed by an inference device having a memory and a processor, The processor: an input information acquisition step of acquiring input information including at least one of workpiece state information indicating the state of the workpiece or die state information indicating the state of the die; an inference step of inferring correction information for correcting the positions of the mold parts based on the input information acquired by the input information acquisition step; Reasoning method.

15. a learning data storage step of storing a plurality of sets of learning data in a learning data storage unit, the learning data being composed of input information including at least one of workpiece state information indicating the state of the workpiece or die state information indicating the state of the die, and correction information for correcting the positions of the die components; a machine learning process of inputting a plurality of sets of the learning data into a learning model to allow the learning model to learn a correlation between the input information and the correction information; a learned model storage step of storing the learned model, which has learned the correlation through the machine learning step, in a learned model storage unit; Equipped with Machine learning methods.

Images