Learned model, management device for injection molding machine, and method for generating training data

Abstract

The invention relates to a learned model, a management device for an injection molding machine, and a method for generating training data. The learning completion model has an input layer, one or more intermediate layers connected to the input layer, and an output layer connected to the intermediate layers, and causes the computer to function as follows: machine learning based on first data and positive solution information is performed; the first data indicates a result detected during the production of the molded article or a result measured after the production of the molded article for each first value set or detected in the predetermined item. The positive solution information indicates, as a positive solution, information relating to the setting of a prescribed item, which is derived from evaluation information indicating the evaluation of a molded article produced when the first value is set or the first value is detected, for each first value. When second data indicating a result detected during production or a result measured after production is input from the input layer, information relating to the setting of the prescribed item is output from the output layer.

Description

Technical Field

[0001] This application claims the benefit of priority based on Japanese Patent Application No. 2023-174295, filed on October 6, 2023. The entire contents of this Japanese patent application are incorporated herein by reference.

[0002] The present invention relates to a learning completed model, a management device for an injection molding machine, and a method for generating training data. Background Art

[0003] Conventionally, accurate production of molded products using injection molding machines required precise settings for the molded products. To determine these settings, the quality of the molded products produced using these settings was verified, and the settings were repeatedly readjusted to derive the correct settings. Consequently, these settings required considerable time and effort on the part of the technicians.

[0004] In recent years, with the improvement of computer processing power, artificial intelligence has been trending towards development. For example, Patent Document 1 proposes a technology for estimating quality through machine learning. As another example, Patent Document 2 proposes a technology that outputs a degree of quality by inputting set values ​​and measured values ​​into a first learning model, and then outputs a set value that reduces the degree of quality by inputting these set values ​​and measured values ​​into a second learning model.

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2021-191619

[0006] Patent Document 2: Japanese Patent Application Laid-Open No. 2021-66084

[0007] However, the technology described in Patent Document 1 uses a learned model to estimate quality based on detection data collected during molding. The quality trend is then evaluated, and corrections to the molding conditions are derived. Specifically, the technology described in Patent Document 1 stops at evaluating molded products using the learned model, and multiple steps are required to derive settings for improving quality based on this evaluation.

[0008] Furthermore, the technique described in Patent Document 2 performs a two-stage process: outputting the quality level and outputting a setting value to reduce the degree of defect based on the quality level. This requires two learning models. Specifically, the technique described in Patent Document 2 complicates the steps and procedures for deriving settings for improving quality. Summary of the Invention

[0009] One embodiment of the present invention provides a technique for reducing the burden of settings for molding a molded product by outputting information related to settings of predetermined items from a learned model.

[0010] A learned model according to one embodiment of the present invention comprises an input layer, one or more intermediate layers connected to the input layer, and an output layer connected to the intermediate layer, so that a computer functions in the following manner: performing machine learning based on first data and correct answer information, wherein the first data represents, for each first value set or detected in a prescribed item for producing a molded product by an injection molding machine, a result detected during the production of the molded product, or a result measured after the production of the molded product, and the correct answer information represents, as a correct answer, information related to the setting of the prescribed item derived from evaluation information representing an evaluation of the molded product produced when the first value is set or when the first value is detected; and outputting, from the output layer, information related to the setting of the prescribed item when second data representing a result detected during production or a result measured after production is input from the input layer.

[0011] Effects of the Invention

[0012] According to one embodiment of the present invention, the burden of setting for performing injection molding is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a diagram showing a state of the injection molding machine according to one embodiment when mold opening is completed.

[0014] Figure 2 It is a diagram showing a state of the injection molding machine according to one embodiment when the mold is clamped.

[0015] Figure 3 This is a schematic diagram showing the cooperation related to the learned model by the learning device and the control device of the injection molding machine according to the first embodiment.

[0016] Figure 4 This is a diagram showing an example of the functional configuration of the learning device and the control device according to the first embodiment.

[0017] Figure 5 This is a diagram illustrating evaluation information acquired by the evaluation acquisition unit according to the first embodiment.

[0018] Figure 6 This is a diagram showing an example of a method for generating a correct answer for the mold clamping force setting value by the correct answer information calculation unit according to the first embodiment.

[0019] Figure 7 This is a diagram illustrating information related to the correct answer generated by the correct answer information calculation unit according to the first embodiment.

[0020] Figure 8 This is a diagram illustrating the difference in mold clamping force acquired by the evaluation acquisition unit according to the first embodiment.

[0021] Figure 9 This is a diagram illustrating the difference in mold clamping force acquired by the evaluation acquisition unit according to the modification.

[0022] Figure 10 This is a diagram illustrating measurement values ​​acquired by the evaluation acquisition unit according to the first embodiment.

[0023] Figure 11 This is a diagram showing an example of a method for calculating information related to a correct answer of a mold clamping force setting value by the correct answer information calculation unit according to the first embodiment.

[0024] Figure 12 This is a diagram illustrating information related to the correct answer calculated by the correct answer information calculation unit according to the first embodiment.

[0025] Figure 13 This is a diagram illustrating the configuration of a learned model used in the learning device and the control device according to the first embodiment.

[0026] Figure 14 This is a diagram illustrating evaluation information acquired by the evaluation acquisition unit according to Modification 2.

[0027] Figure 15 This is a graph showing the correspondence between the pressure holding time and the weight according to Modification 2.

[0028] Figure 16 This is a diagram illustrating information related to the correct answer generated by the correct answer information calculation unit according to the embodiment of Modification 2.

[0029] Figure 17 This is a diagram illustrating the configuration of a learning device, a group management device, and an injection molding machine according to a second embodiment.

[0030] Explanation of symbols

[0031] 10-injection molding machine, 700-control device, 701-CPU, 702-storage medium, 711-communication control unit, 712-update unit, 713-injection molding processing unit, 714-setting unit, 715-inference unit, 721-learned model storage unit, 1300-learning device, 1301-CPU, 1302-storage medium, 1303-communication interface, 1311-acquisition unit, 1312-evaluation acquisition unit, 1313-correct answer information calculation unit, 1314-training data generation unit, 1315-learning unit, 1316-communication control unit, 1321-training data storage unit, 1322-learned model storage unit, 2700-group management device. DETAILED DESCRIPTION

[0032] The following describes embodiments of the present invention with reference to the accompanying drawings. The embodiments described below are illustrative rather than limiting, and all features and combinations thereof described in the embodiments are not necessarily essential to the invention. In addition, in the drawings, identical or corresponding components may be denoted by identical or corresponding reference numerals, and their description may be omitted.

[0033] Figure 1 This is a diagram showing a state of the injection molding machine according to the first embodiment when mold opening is completed. Figure 2 This figure shows the injection molding machine according to the first embodiment during mold clamping. In this specification, the X-axis, Y-axis, and Z-axis directions are perpendicular to each other. The X-axis and Y-axis directions represent horizontal directions, while the Z-axis direction represents the vertical direction. When the mold clamping device 100 is horizontal, the X-axis direction represents the mold opening and closing direction, and the Y-axis direction represents the width of the injection molding machine 10. The negative side in the Y-axis direction is referred to as the operating side, and the positive side in the Y-axis direction is referred to as the side opposite the operating side.

[0034] like Figures 1 and 2 As shown, the injection molding machine 10 includes a mold clamping unit 100 for opening and closing the mold unit 800; an ejection unit 200 for ejecting a molded product formed by the mold unit 800; an injection unit 300 for injecting molding material into the mold unit 800; a moving unit 400 for moving the injection unit 300 forward and backward relative to the mold unit 800; a control unit 700 for controlling the various components of the injection molding machine 10; and a frame 900 for supporting the various components of the injection molding machine 10. The frame 900 includes a mold clamping unit frame 910 that supports the mold clamping unit 100 and an injection unit frame 920 that supports the injection unit 300. The mold clamping unit frame 910 and the injection unit frame 920 are each mounted on the base plate 2 via leveling casters 930. The control unit 700 is located within the interior space of the injection unit frame 920. The components of the injection molding machine 10 are described below.

[0035] (Mold clamping device)

[0036] In the description of the mold clamping device 100, the moving direction of the movable platen 120 when the mold is closed (for example, the positive direction of the X axis) is set as the front, and the moving direction of the movable platen 120 when the mold is open (for example, the negative direction of the X axis) is set as the rear.

[0037] The mold clamping device 100 performs mold closing, pressurization, clamping, depressurization, and mold opening of the mold assembly 800. The mold assembly 800 includes a fixed mold 810 and a movable mold 820. The mold clamping device 100 is, for example, horizontal, with the mold opening and closing direction being horizontal. The mold clamping device 100 includes a fixed platen 110 to which the fixed mold 810 is mounted, a movable platen 120 to which the movable mold 820 is mounted, and a moving mechanism 102 that moves the movable platen 120 relative to the fixed platen 110 in the mold opening and closing direction.

[0038] The fixed platen 110 is fixed to the mold clamping device frame 910. The fixed mold 810 is attached to the surface of the fixed platen 110 that faces the movable platen 120.

[0039] The movable platen 120 is arranged to be movable in the mold opening and closing direction relative to the mold clamping device frame 910. A guide 101 for guiding the movable platen 120 is laid on the mold clamping device frame 910. The movable mold 820 is mounted on the surface of the movable platen 120 facing the fixed platen 110.

[0040] The moving mechanism 102 moves the movable platen 120 forward and backward relative to the fixed platen 110, thereby closing, increasing pressure, clamping, releasing pressure, and opening the mold assembly 800. The moving mechanism 102 includes a toggle base 130 spaced apart from the fixed platen 110, a connecting rod 140 connecting the fixed platen 110 and the toggle base 130, a toggle mechanism 150 that moves the movable platen 120 relative to the toggle base 130 in the mold opening and closing direction, a mold clamping motor 160 that operates the toggle mechanism 150, a motion conversion mechanism 170 that converts the rotational motion of the mold clamping motor 160 into linear motion, and a mold thickness adjustment mechanism 180 that adjusts the gap between the fixed platen 110 and the toggle base 130.

[0041] The toggle base 130 is spaced apart from the fixed platen 110 and mounted on the mold clamping unit frame 910 so as to be movable in the mold opening and closing direction. Furthermore, the toggle base 130 can be configured to be movable along a guide provided on the mold clamping unit frame 910. The guide of the toggle base 130 can be the same as the guide 101 of the movable platen 120.

[0042] In addition, in this embodiment, the fixed pressure plate 110 is fixed relative to the mold clamping device frame 910, and the elbow seat 130 is configured to be freely movable relative to the mold clamping device frame 910 in the mold opening and closing direction, but the elbow seat 130 may also be fixed relative to the mold clamping device frame 910, and the fixed pressure plate 110 may be configured to be freely movable relative to the mold clamping device frame 910 in the mold opening and closing direction.

[0043] The connecting rod 140 connects the fixed platen 110 and the toggle seat 130 at a distance L in the mold opening and closing direction. A plurality of connecting rods 140 (e.g., four) may be used. The plurality of connecting rods 140 are arranged parallel to the mold opening and closing direction and extend according to the mold clamping force. A connecting rod strain detector 141 for detecting the strain of the connecting rod 140 may be provided on at least one of the connecting rods 140. The connecting rod strain detector 141 sends a signal indicating its detection result to the control device 700. The detection result of the connecting rod strain detector 141 is used for detecting the mold clamping force, etc.

[0044] In this embodiment, the tie-bar strain detector 141 is used as the mold clamping force detector for detecting the mold clamping force. However, the present invention is not limited to this type. The mold clamping force detector is not limited to a strain gauge type; it may also be a piezoelectric, capacitive, hydraulic, or electromagnetic type. Its mounting location is not limited to the tie-bar 140.

[0045] The toggle mechanism 150 is disposed between the movable platen 120 and the toggle base 130 and moves the movable platen 120 relative to the toggle base 130 in the mold opening and closing direction. The toggle mechanism 150 includes a crosshead 151 that moves in the mold opening and closing direction and a pair of link assemblies that flex and extend as the crosshead 151 moves. The pair of link assemblies each includes a first link 152 and a second link 153, each connected by a pin or the like so as to be able to flex and extend freely. The first link 152 is mounted by a pin or the like so as to be able to swing freely relative to the movable platen 120. The second link 153 is mounted by a pin or the like so as to be able to swing freely relative to the toggle base 130. The second link 153 is mounted to the crosshead 151 via a third link 154. When the crosshead 151 is moved forward or backward relative to the toggle base 130, the first link 152 and the second link 153 flex and extend, causing the movable platen 120 to advance or retract relative to the toggle base 130.

[0046] In addition, the structure of the toggle mechanism 150 is not limited to Figure 1 and Figure 2 For example, in Figure 1 and Figure 2 In the embodiment, the number of nodes of each link group is five, but it may be four, or one end of the third link 154 may be connected to the node between the first link 152 and the second link 153 .

[0047] The mold clamping motor 160 is mounted on the toggle base 130 and operates the toggle mechanism 150. The mold clamping motor 160 advances and retracts the crosshead 151 relative to the toggle base 130, thereby flexing and extending the first link 152 and the second link 153, thereby advancing and retracting the movable platen 120 relative to the toggle base 130. The mold clamping motor 160 is directly connected to the motion conversion mechanism 170, but may alternatively be connected via a belt, pulley, or the like.

[0048] The motion conversion mechanism 170 converts the rotational motion of the mold clamping motor 160 into the linear motion of the crosshead 151. The motion conversion mechanism 170 includes a screw shaft and a screw nut threadedly engaged with the screw shaft. Balls or rollers may be interposed between the screw shaft and the screw nut.

[0049] The mold clamping device 100 performs a mold closing process, a pressure increasing process, a mold clamping process, a pressure releasing process, a mold opening process, and the like under the control of the control device 700 .

[0050] During the mold closing process, the mold clamping motor 160 is driven to advance the crosshead 151 at a set movement speed to the mold closing end position, causing the movable platen 120 to advance, bringing the movable mold 820 into contact with the fixed mold 810. The position and movement speed of the crosshead 151 are detected, for example, using a mold clamping motor encoder 161. The mold clamping motor encoder 161 detects the rotation of the mold clamping motor 160 and transmits a signal indicating the detection result to the control device 700.

[0051] The crosshead position detector for detecting the position of the crosshead 151 and the crosshead moving speed detector for detecting the moving speed of the crosshead 151 are not limited to the mold clamping motor encoder 161; conventional detectors can be used. Furthermore, the movable platen position detector for detecting the position of the movable platen 120 and the movable platen moving speed detector for detecting the moving speed of the movable platen 120 are not limited to the mold clamping motor encoder 161; conventional detectors can be used.

[0052] In the pressure increasing step, the mold clamping motor 160 is further driven to further advance the crosshead 151 from the mold closing end position to the mold clamping position, thereby generating a mold clamping force.

[0053] In the mold closing process, the mold closing motor 160 is driven to maintain the position of the cross head 151 at the mold closing position. In the mold closing process, the mold closing force generated in the pressure increasing process is maintained. In the mold closing process, a cavity space 801 is formed between the movable mold 820 and the fixed mold 810 (refer to Figure 2 ), the injection device 300 fills the cavity space 801 with liquid molding material. The filled molding material solidifies, thereby obtaining a molded product.

[0054] The number of mold cavities 801 can be one or more. In the latter case, multiple molded products can be obtained simultaneously. An insert can be placed in a portion of the mold cavity 801, while another portion of the mold cavity 801 is filled with molding material. This can produce a molded product in which the insert and molding material are integrated.

[0055] During the depressurization process, the mold clamping motor 160 is driven to retract the crosshead 151 from the mold clamping position to the mold opening start position, thereby retracting the movable platen 120 to reduce the mold clamping force. The mold opening start position and the mold closing end position may be the same position.

[0056] During the mold opening process, the mold clamping motor 160 is driven to retract the crosshead 151 from the mold opening start position to the mold opening completion position at a set movement speed, causing the movable platen 120 to retract, separating the movable mold 820 from the fixed mold 810. The ejector device 200 then ejects the molded product from the movable mold 820.

[0057] The setting conditions in the mold closing process, the pressure increasing process and the mold clamping process are uniformly set as a series of setting conditions. For example, the moving speed, position (including the mold closing start position, the moving speed switching position, the mold closing end position and the mold clamping position) and the mold clamping force of the crosshead 151 in the mold closing process and the pressure increasing process are uniformly set as a series of setting conditions. The mold closing start position, the moving speed switching position, the mold closing end position and the mold clamping position are arranged in sequence from the back to the front, and represent the starting point and end point of the interval for setting the moving speed. The moving speed is set for each interval. The moving speed switching position can be one or more. The moving speed switching position can also be not set. It is also possible to set only one of the mold clamping position and the mold clamping force.

[0058] The setting conditions in the depressurization process and the mold opening process are also set in the same way. For example, the moving speed and position (mold opening start position, moving speed switching position and mold opening completion position) of the crosshead 151 in the depressurization process and the mold opening process are uniformly set as a series of setting conditions. The mold opening start position, the moving speed switching position and the mold opening completion position are arranged in sequence from the front to the back, and represent the starting point and end point of the interval for setting the moving speed. The moving speed is set for each interval. The moving speed switching position can be one or more. The moving speed switching position can also be not set. The mold opening start position and the mold closing end position can be the same position. In addition, the mold opening completion position and the mold closing start position can be the same position.

[0059] Alternatively, the moving speed and position of the movable platen 120 may be set instead of the moving speed and position of the crosshead 151. Furthermore, the mold clamping force may be set instead of the position of the crosshead (e.g., mold clamping position) or the position of the movable platen.

[0060] The toggle mechanism 150 amplifies the driving force of the mold clamping motor 160 and transmits it to the movable platen 120. This amplification factor is also called the toggle magnification. The toggle magnification changes depending on the angle θ (hereinafter referred to as "link angle θ") formed between the first link 152 and the second link 153. Link angle θ is determined by the position of the crosshead 151. The toggle magnification is maximum when link angle θ is 180°.

[0061] If the thickness of the mold assembly 800 changes due to, for example, replacement of the mold assembly 800 or temperature fluctuations, mold thickness adjustment is performed to achieve a predetermined clamping force during mold clamping. During mold thickness adjustment, for example, the gap L between the fixed platen 110 and the toggle base 130 is adjusted so that the link angle θ of the toggle mechanism 150 reaches a predetermined angle when the movable mold 820 contacts the fixed mold 810.

[0062] The mold clamping device 100 includes a mold thickness adjustment mechanism 180. The mold thickness adjustment mechanism 180 adjusts the distance L between the fixed platen 110 and the toggle base 130, thereby adjusting the mold thickness. Mold thickness adjustment is performed, for example, between the end of a molding cycle and the start of the next molding cycle. The mold thickness adjustment mechanism 180 includes, for example, a screw shaft 181 formed at the rear end of the connecting rod 140; a screw nut 182 held in a rotatable and non-retractable manner by the toggle base 130; and a mold thickness adjustment motor 183 that rotates the screw nut 182, which is threadedly engaged with the screw shaft 181.

[0063] A screw shaft 181 and a screw nut 182 are provided for each connecting rod 140. The rotational driving force of the mold thickness adjustment motor 183 can be transmitted to the multiple screw nuts 182 via the rotational driving force transmission unit 185. The multiple screw nuts 182 can be rotated synchronously. In addition, by changing the transmission path of the rotational driving force transmission unit 185, the multiple screw nuts 182 can also be rotated independently.

[0064] The rotational driving force transmission unit 185 is composed of, for example, gears. In this case, driven gears are formed on the outer periphery of each lead screw nut 182, a driving gear is attached to the output shaft of the mold thickness adjustment motor 183, and an intermediate gear meshing with the multiple driven gears and the driving gear is rotatably held in the center of the toggle base 130. Alternatively, the rotational driving force transmission unit 185 may be composed of a belt, pulley, or the like, instead of gears.

[0065] The operation of the mold thickness adjustment mechanism 180 is controlled by the control device 700. The control device 700 drives the mold thickness adjustment motor 183 to rotate the lead screw nut 182. As a result, the position of the toggle base 130 relative to the connecting rod 140 is adjusted, and the distance L between the fixed platen 110 and the toggle base 130 is adjusted. Alternatively, multiple mold thickness adjustment mechanisms may be used in combination.

[0066] The gap L is detected using the mold thickness adjustment motor encoder 184. The mold thickness adjustment motor encoder 184 detects the rotation amount and rotation direction of the mold thickness adjustment motor 183 and transmits a signal indicating the detection result to the control device 700. The detection result of the mold thickness adjustment motor encoder 184 is used to monitor and control the position of the toggle base 130 and the gap L. The toggle base position detector for detecting the position of the toggle base 130 and the gap detector for detecting the gap L are not limited to the mold thickness adjustment motor encoder 184; conventional detectors can be used.

[0067] The mold clamping device 100 may include a mold temperature regulator for regulating the temperature of the mold device 800. The mold device 800 has a flow path for a temperature regulating medium therein. The mold temperature regulator regulates the temperature of the temperature regulating medium supplied to the flow path of the mold device 800, thereby regulating the temperature of the mold device 800.

[0068] Furthermore, the mold clamping device 100 of the present embodiment is a horizontal type in which the mold opening and closing direction is the horizontal direction, but may be a vertical type in which the mold opening and closing direction is the vertical direction.

[0069] The mold clamping device 100 of this embodiment includes the mold clamping motor 160 as a driving source, but may include a hydraulic cylinder instead of the mold clamping motor 160. Furthermore, the mold clamping device 100 may include a linear motor for mold opening and closing or an electromagnet for mold clamping.

[0070] (Ejector device)

[0071] In the description of the ejector device 200, similar to the description of the mold clamping device 100, etc., the moving direction of the movable platen 120 when the mold is closed (for example, the positive direction of the X-axis) is set to the front, and the moving direction of the movable platen 120 when the mold is open (for example, the negative direction of the X-axis) is set to the rear.

[0072] The ejector device 200 is mounted on the movable platen 120 and moves forward and backward along with the movable platen 120. The ejector device 200 includes an ejector rod 210 for ejecting the molded product from the mold assembly 800, and a drive mechanism 220 for moving the ejector rod 210 in the direction of movement of the movable platen 120 (the X-axis direction).

[0073] The ejector rod 210 is arranged to be able to move forward and backward in the through hole of the movable platen 120. The front end of the ejector rod 210 contacts the ejector plate 826 of the movable mold 820. The front end of the ejector rod 210 may or may not be connected to the ejector plate 826.

[0074] The drive mechanism 220 includes, for example, an ejector motor and a motion conversion mechanism that converts the motor's rotational motion into linear motion of the ejector rod 210. The motion conversion mechanism includes a screw shaft and a screw nut threadedly engaged with the screw shaft. Balls or rollers may be interposed between the screw shaft and the screw nut.

[0075] The ejector device 200 performs the ejection process under the control of the control device 700. During the ejection process, the ejector rod 210 advances from the standby position to the ejection position at a set speed, advancing the ejector plate 826 to eject the molded product. The ejector motor then drives the ejector rod 210 backward at the set speed, retracting the ejector plate 826 to its original standby position.

[0076] For example, an ejector motor encoder is used to detect the position and movement speed of the ejector rod 210. The ejector motor encoder detects the rotation of the ejector motor and transmits a signal indicating the detection result to the control device 700. The ejector rod position detector that detects the position of the ejector rod 210 and the ejector rod movement speed detector that detects the movement speed of the ejector rod 210 are not limited to the ejector motor encoder; conventional detectors can be used.

[0077] (Injection device)

[0078] In the description of the injection device 300, unlike the description of the mold clamping device 100 and the description of the ejection device 200, the movement direction of the screw 330 during filling (for example, the negative direction of the X-axis) is set to the front, and the movement direction of the screw 330 during metering (for example, the positive direction of the X-axis) is set to the rear.

[0079] The injection device 300 is mounted on a sliding base 301, which is configured to be freely movable relative to the injection device frame 920. The injection device 300 is configured to be freely movable relative to the mold device 800. The injection device 300 contacts the mold device 800 and fills the cavity space 801 within the mold device 800 with the molding material measured within the cylinder 310. The injection device 300 includes, for example, a cylinder 310 for heating the molding material, a nozzle 320 provided at the front end of the cylinder 310, a screw 330 configured to be freely movable and rotatable within the cylinder 310, a metering motor 340 for rotating the screw 330, an injection motor 350 for moving the screw 330 forward and backward, and a load detector 360 for detecting the load transmitted between the injection motor 350 and the screw 330.

[0080] Cylinder 310 heats the molding material supplied into it from supply port 311. The molding material, for example, includes resin. The molding material is formed into pellets, for example, and supplied to supply port 311 in a solid state. Supply port 311 is formed at the rear of cylinder 310. A cooler 312, such as a water-cooled cylinder, is provided on the outer periphery of the rear portion of cylinder 310. A heater 313, such as a belt heater, and a temperature detector 314 are provided on the outer periphery of cylinder 310, further in front of cooler 312.

[0081] The cylinder 310 is divided into a plurality of zones along the axial direction (e.g., the X-axis direction) of the cylinder 310. Heaters 313 and temperature detectors 314 are provided in each of the zones. A set temperature is set for each of the zones, and the control device 700 controls the heater 313 so that the temperature detected by the temperature detector 314 reaches the set temperature.

[0082] The nozzle 320 is provided at the front end of the cylinder 310 and presses the mold device 800. A heater 313 and a temperature detector 314 are provided on the periphery of the nozzle 320. The control device 700 controls the heater 313 so that the detected temperature of the nozzle 320 becomes a set temperature.

[0083] Screw 330 is configured to rotate freely and move forward and backward within cylinder 310. When screw 330 is rotated, the molding material is conveyed forward along the spiral groove of screw 330. As the molding material is conveyed forward, it is gradually melted by the heat from cylinder 310. As the liquid molding material is conveyed forward of screw 330 and accumulates in the front portion of cylinder 310, screw 330 is retracted. Then, when screw 330 is advanced, the liquid molding material accumulated in front of screw 330 is injected from nozzle 320 and fills mold device 800.

[0084] The check ring 331 is installed at the front of the screw 330 so as to be able to move forward and backward. The check ring 331 acts as a check valve to prevent the molding material from flowing backward from the front to the rear of the screw 330 when the screw 330 is pushed forward.

[0085] When the screw 330 is advanced, the check ring 331 is pushed backward by the pressure of the molding material in front of the screw 330, and moves back relative to the screw 330 to a closed position that blocks the flow path of the molding material (see FIG. Figure 2 ). This prevents the molding material accumulated in front of the screw 330 from flowing backward.

[0086] On the other hand, when the screw 330 is rotated, the check ring 331 is pushed forward by the pressure of the molding material conveyed forward along the spiral groove of the screw 330, and moves relative to the screw 330 to an open position (refer to FIG. Figure 1). As a result, the molding material is transported to the front of the screw 330.

[0087] The check ring 331 may be of a co-rotating type that rotates together with the screw 330 or a non-co-rotating type that does not rotate together with the screw 330 .

[0088] In addition, the injection device 300 may include a drive source that moves the check ring 331 forward and backward relative to the screw 330 between the open position and the closed position.

[0089] The metering motor 340 rotates the screw 330. The driving source for rotating the screw 330 is not limited to the metering motor 340, and may be, for example, a hydraulic pump.

[0090] The injection motor 350 advances and retracts the screw 330. A motion conversion mechanism, for example, is provided between the injection motor 350 and the screw 330 to convert the rotational motion of the injection motor 350 into linear motion of the screw 330. The motion conversion mechanism, for example, includes a screw shaft and a screw nut threadedly engaged with the screw shaft. Balls, rollers, or the like may be provided between the screw shaft and the screw nut. The drive source for advancing and retracting the screw 330 is not limited to the injection motor 350; for example, a hydraulic cylinder may be provided.

[0091] The load detector 360 detects the load transmitted between the injection motor 350 and the screw 330. The detected load is converted into pressure by the control device 700. The load detector 360 is provided in the load transmission path between the injection motor 350 and the screw 330, and detects the load acting on the load detector 360.

[0092] The load detector 360 transmits a signal of the detected load to the control device 700. The load detected by the load detector 360 is converted into a pressure acting between the screw 330 and the molding material, and is used to control and monitor the pressure exerted on the screw 330 by the molding material, the back pressure on the screw 330, and the pressure acting on the molding material from the screw 330.

[0093] The pressure detector for detecting the pressure of the molding material is not limited to the load detector 360 , and a conventional detector can be used. For example, a nozzle pressure sensor or a mold internal pressure sensor can be used. The nozzle pressure sensor is provided at the nozzle 320 .

[0094] The injection device 300 performs a metering process, a filling process, a pressure-maintaining process, etc. under the control of the control device 700. The filling process and the pressure-maintaining process can be collectively referred to as an injection process.

[0095] In the metering process, the metering motor 340 is driven to rotate the screw 330 at a set speed, and the molding material is transported to the front along the spiral groove of the screw 330. As a result, the molding material is gradually melted. As the liquid molding material is transported to the front of the screw 330 and accumulates at the front of the cylinder 310, the screw 330 is retracted. For example, the rotation speed of the screw 330 is detected using a metering motor encoder 341. The metering motor encoder 341 detects the rotation of the metering motor 340 and sends a signal indicating its detection result to the control device 700. In addition, the screw speed detector that detects the rotation speed of the screw 330 is not limited to the metering motor encoder 341, and a conventional detector can be used.

[0096] During the metering process, to prevent the screw 330 from rapidly retreating, the injection motor 350 can be driven to apply a set back pressure to the screw 330. For example, the back pressure on the screw 330 can be detected using a load detector 360. The metering process ends when the screw 330 retreats to the metering end position and a predetermined amount of molding material accumulates in front of the screw 330.

[0097] The position and speed of the screw 330 during the metering process are uniformly set as a series of setting conditions. For example, the metering start position, speed switching position, and metering end position are set. These positions are arranged in sequence from the front to the back and represent the start and end points of the speed setting interval. The speed is set for each interval. There can be one or more speed switching positions. It is also possible to set no speed switching position. In addition, the back pressure can be set for each interval.

[0098] In the filling process, the injection motor 350 is driven to make the screw 330 move forward at a set moving speed, and the liquid molding material accumulated in front of the screw 330 is filled into the cavity space 801 in the mold device 800. For example, the position and moving speed of the screw 330 are detected using the injection motor encoder 351. The injection motor encoder 351 detects the rotation of the injection motor 350 and sends a signal indicating its detection result to the control device 700. If the position of the screw 330 reaches the set position, the switch from the filling process to the pressure holding process is performed (the so-called V / P switching). The position where the V / P switching is performed is also called the V / P switching position. The set moving speed of the screw 330 can be changed according to the position of the screw 330, time, etc.

[0099] The position and movement speed of screw 330 during the filling process are uniformly set as a series of setting conditions. For example, the filling start position (also called the "injection start position"), the movement speed switching position, and the V / P switching position are set. These positions are arranged in sequence from the back to the front and represent the starting and ending points of the interval for setting the movement speed. The movement speed is set for each interval. There can be one or more movement speed switching positions. It is also possible not to set a movement speed switching position.

[0100] An upper limit value for the pressure of the screw 330 is set for each interval of the screw 330's travel speed. The pressure of the screw 330 is detected by a load detector 360. When the pressure of the screw 330 is below the set pressure, the screw 330 advances at the set travel speed. On the other hand, when the pressure of the screw 330 exceeds the set pressure, the screw 330 advances at a speed slower than the set travel speed to reduce the pressure of the screw 330 to below the set pressure in order to protect the mold.

[0101] Furthermore, during the filling process, after the screw 330 reaches the V / P switching position, the screw 330 can be paused at the V / P switching position before the V / P switching is performed. Alternatively, instead of stopping the screw 330 immediately before the V / P switching, the screw 330 can be moved forward or backward at a slight speed. Furthermore, the screw position detector for detecting the position of the screw 330 and the screw movement speed detector for detecting the movement speed of the screw 330 are not limited to the injection motor encoder 351; conventional detectors can be used.

[0102] In the holding pressure process, the injection motor 350 is driven to push the screw 330 forward, and the pressure of the molding material at the front end of the screw 330 (hereinafter also referred to as "holding pressure") is maintained at the set pressure, and the molding material remaining in the cylinder 310 is pushed toward the mold device 800. The insufficient amount of molding material in the mold device 800 due to cooling shrinkage can be supplemented. For example, the holding pressure is detected using a load detector 360. The setting value of the holding pressure can be changed according to the time elapsed since the start of the holding pressure process. The holding pressure and the holding time of the holding pressure in multiple holding pressure processes can be set separately, or they can be set uniformly as a series of setting conditions.

[0103] During the pressure holding process, the molding material in the cavity space 801 within the mold device 800 gradually cools. At the end of the pressure holding process, the inlet of the cavity space 801 is blocked by the solidified molding material. This state is called gate sealing and prevents the molding material from flowing back out of the cavity space 801. After the pressure holding process, the cooling process begins. During the cooling process, the molding material in the cavity space 801 solidifies. To shorten the molding cycle time, a metering process can be performed during the cooling process.

[0104] While the injection device 300 of this embodiment employs a coaxial screw system, it may also employ a pre-molding system. Pre-molding system injection devices supply the molding material melted in a plasticizing cylinder to an injection cylinder, which then injects the molding material into the mold assembly. Within the plasticizing cylinder, the screw is configured to rotate freely and not advance or retreat, or to rotate freely and advance and retreat freely. Meanwhile, within the injection cylinder, the plunger is configured to advance and retreat freely.

[0105] Furthermore, while the injection unit 300 of this embodiment is a horizontal type with the axial direction of the cylinder 310 being horizontal, it may also be a vertical type with the axial direction of the cylinder 310 being vertical. The mold clamping unit used in conjunction with the vertical injection unit 300 may be either vertical or horizontal. Similarly, the mold clamping unit used in conjunction with the horizontal injection unit 300 may be either horizontal or vertical.

[0106] (Mobile device)

[0107] In the description of the moving device 400, similar to the description of the injection device 300, the moving direction of the screw 330 during filling (e.g., the negative direction of the X-axis) is set to the front, and the moving direction of the screw 330 during metering (e.g., the positive direction of the X-axis) is set to the rear.

[0108] The moving device 400 moves the injection device 300 forward and backward relative to the mold device 800. Furthermore, the moving device 400 presses the nozzle 320 against the mold device 800, generating nozzle contact pressure. The moving device 400 includes a hydraulic pump 410, a motor 420 as a drive source, and a hydraulic cylinder 430 as a hydraulic actuator.

[0109] The hydraulic pump 410 has a first port 411 and a second port 412. The hydraulic pump 410 is a bidirectionally rotatable pump. By switching the rotational direction of the motor 420, the hydraulic pump 410 draws hydraulic fluid (e.g., oil) from one of the first and second ports 411, 412 and discharges it from the other port, generating hydraulic pressure. Alternatively, the hydraulic pump 410 can draw hydraulic fluid from a tank and discharge it from either the first or second port 411, 412.

[0110] The motor 420 operates the hydraulic pump 410. The motor 420 drives the hydraulic pump 410 with a rotation direction and torque corresponding to a control signal from the control device 700. The motor 420 may be an electric motor or an electric servomotor.

[0111] The hydraulic cylinder 430 includes a cylinder body 431, a piston 432, and a piston rod 433. The cylinder body 431 is fixed to the injection device 300. The piston 432 divides the interior of the cylinder body 431 into a front chamber 435 serving as a first chamber and a rear chamber 436 serving as a second chamber. The piston rod 433 is fixed to the fixed platen 110.

[0112] The front chamber 435 of the hydraulic cylinder 430 is connected to the first port 411 of the hydraulic pump 410 via the first flow path 401. The hydraulic fluid discharged from the first port 411 is supplied to the front chamber 435 via the first flow path 401, thereby pushing the injection device 300 forward. The injection device 300 advances, pressing the nozzle 320 against the fixed mold 810. The front chamber 435 functions as a pressure chamber, generating nozzle contact pressure for the nozzle 320 using the pressure of the hydraulic fluid supplied from the hydraulic pump 410.

[0113] Meanwhile, the rear chamber 436 of the hydraulic cylinder 430 is connected to the second port 412 of the hydraulic pump 410 via the second flow path 402. The hydraulic fluid discharged from the second port 412 is supplied to the rear chamber 436 of the hydraulic cylinder 430 via the second flow path 402, thereby pushing the injection device 300 rearward. The injection device 300 retracts, and the nozzle 320 separates from the fixed mold 810.

[0114] In this embodiment, the moving device 400 includes the hydraulic cylinder 430 , but the present invention is not limited thereto. For example, instead of the hydraulic cylinder 430 , an electric motor and a motion conversion mechanism that converts the rotational motion of the electric motor into linear motion of the injection device 300 may be used.

[0115] (Control device)

[0116] The control device 700 is constituted by a computer, for example. Figures 1 and 2 As shown, the control device 700 includes a CPU (Central Processing Unit) 701, a storage medium such as a memory 702, an input interface 703, an output interface 704, and a communication interface 705. The control device 700 performs various controls by causing the CPU 701 to execute programs stored in the storage medium 702. Furthermore, the control device 700 receives external signals via the input interface 703 and transmits signals to the outside via the output interface 704.

[0117] The control device 700 repeatedly produces molded products by performing the metering process, mold closing process, pressure increase process, mold clamping process, filling process, pressure holding process, cooling process, pressure release process, mold opening process, and ejection process. The series of operations required to produce a molded product, such as the process from the start of the metering process to the start of the next metering process, is also referred to as a "shot" or "molding cycle." Furthermore, the time required for one shot is also referred to as the "molding cycle time" or "cycle time."

[0118] A primary molding cycle, for example, sequentially includes a metering step, a mold closing step, a pressure increase step, a mold closing step, a filling step, a pressure holding step, a cooling step, a pressure release step, a mold opening step, and an ejection step. The order used here refers to the order in which each step begins. The filling step, the pressure holding step, and the cooling step occur during the mold closing step. Alternatively, the start of the mold closing step can coincide with the start of the filling step. The end of the pressure release step coincides with the start of the mold opening step.

[0119] In addition, in order to shorten the molding cycle time, multiple processes can be carried out simultaneously. For example, the metering process can be carried out during the cooling process of the previous molding cycle, or during the mold closing process. In this case, the mold closing process can be set to be carried out at the beginning of the molding cycle. In addition, the filling process can be started during the mold closing process. In addition, the ejection process can be started during the mold opening process. When an on-off valve is set to open and close the flow path of the nozzle 320, the mold opening process can be started during the metering process. Because even if the mold opening process is started during the metering process, as long as the on-off valve closes the flow path of the nozzle 320, the molding material will not leak from the nozzle 320.

[0120] In addition, a primary molding cycle may include processes other than the metering step, mold closing step, pressure increasing step, mold clamping step, filling step, pressure holding step, cooling step, pressure releasing step, mold opening step, and ejection step.

[0121] For example, a pre-metering backdraft process can be performed after the pressure holding process is completed and before the metering process begins, in which the screw 330 is retracted to a preset metering start position. This can reduce the pressure of the molding material accumulated in front of the screw 330 before the metering process begins, thereby preventing the screw 330 from rapidly retracting when the metering process begins.

[0122] Furthermore, a post-metering backsucking step may be performed after the metering step is completed and before the filling step begins, in which the screw 330 is retracted to a preset filling start position (also referred to as the "injection start position"). This can reduce the pressure of the molding material accumulated in front of the screw 330 before the filling step begins, thereby preventing leakage of the molding material from the nozzle 320 before the filling step begins.

[0123] The control device 700 is connected to an operating device 750 that receives user input operations and a display device 760 that displays a screen. The operating device 750 and the display device 760 are composed of, for example, a touch panel 770 and may be integrated. The touch panel 770, serving as the display device 760, displays a screen under the control of the control device 700. Information such as the settings of the injection molding machine 10 and the current status of the injection molding machine 10 can be displayed on the screen of the touch panel 770. The touch panel 770 can receive operations in the displayed screen area. Furthermore, operating components such as buttons and input fields for receiving user input operations can be displayed on the screen area of ​​the touch panel 770. The touch panel 770, serving as the operating device 750, detects user input operations on the screen and outputs signals corresponding to the input operations to the control device 700. Thus, for example, a user can operate the operating components provided on the screen while confirming information displayed on the screen to configure the injection molding machine 10 (including inputting set values). Furthermore, by operating the operating components provided on the screen, the user can cause the injection molding machine 10 to operate in accordance with the corresponding operating components. The operation of the injection molding machine 10 may include, for example, the operation (including stopping) of the mold clamping device 100, the ejection device 200, the injection device 300, the moving device 400, etc. Furthermore, the operation of the injection molding machine 10 may include switching of the screen displayed on the touch panel 770 serving as the display device 760.

[0124] In this embodiment, the operating device 750 and the display device 760 are described as being integrated into a touch panel 770. However, they may be provided separately. Furthermore, multiple operating devices 750 may be provided. The operating device 750 and the display device 760 are disposed on the operating side (in the negative Y-axis direction) of the mold clamping unit 100 (more specifically, the fixed platen 110).

[0125] (First embodiment)

[0126] When performing molding with an injection molding machine, there is a desire to use a learned model for setting. In this case, machine learning is required to generate the learned model.

[0127] The injection molding machine could also be considered for the learning phase of machine learning. However, this would require a high level of processing power from the injection molding machine's processing equipment. Therefore, it could be considered to use a separate information processing device to perform the learning process.

[0128] Therefore, this embodiment describes an example in which the learning phase and the inference phase are performed in separate devices. Specifically, a learned model created by a learning device during the learning phase is mounted on an injection molding machine. Furthermore, while this embodiment describes an example in which the learning phase and the inference phase are performed in separate devices, the method is not limited to performing the learning phase and the inference phase in separate devices; the learning phase and the inference phase can also be performed in the same device.

[0129] Figure 3 Schematic diagram showing the cooperation between the learning device and the injection molding machine control device according to the present embodiment with respect to the learned model.

[0130] exist Figure 3 In the illustrated example, a test injection molding machine 1350 , a learning device 1300 , and a control device 700 for the injection molding machine 10 are shown.

[0131] exist Figure 3 In the example shown, the test injection molding machine 1350 and the learning device 1300 can be considered to be owned by, for example, the manufacturer of the injection molding machine 10, but are not limited to being owned by the manufacturer of the injection molding machine. For example, they can be owned by the shipping destination of the injection molding machine, or they can be owned by the service provider who generates the learned model. Figure 3 The injection molding machine 10 and the control device 700 shown may also show the status before shipment, for example.

[0132] The test injection molding machine 1350 is an injection molding machine used for machine learning. The structure of the test injection molding machine 1350 is the same as that of the injection molding machine 10, and therefore, description thereof will be omitted.

[0133] The test injection molding machine 1350 according to this embodiment produces molded products according to user settings. Furthermore, the test injection molding machine 1350 outputs waveform data measured during injection molding to the learning device 1300. In this embodiment, data representing the detection results of sensors installed in the test injection molding machine 1350 or the injection molding machine 10 in a time series is referred to as waveform data.

[0134] The learning device 1300 generates training data based on the input waveform data and stores the training data in the training data storage unit 1321 .

[0135] However, in order to apply artificial intelligence, a learning (training) phase and an inference (inference) phase need to be performed.

[0136] The learning device 1300 executes a learning phase. Therefore, the learning unit 1315 of the learning device 1300 causes the neural network to read the training data stored in the training data storage unit 1321 and generates a network with adjusted synaptic weights and biases as a learned model LM.

[0137] For example, the learning device 1300 reads a large amount of training data and performs machine learning based on back propagation using a neural network, thereby generating a learned model LM.

[0138] Learning device 1300 can be, for example, a local server installed in a factory or a cloud server. Furthermore, learning device 1300 can be a stationary terminal device or a portable terminal device (mobile terminal) installed in a factory or the like. Stationary terminal devices include, for example, desktop PCs (personal computers). Mobile terminal devices include, for example, smartphones, tablets, and portable PCs.

[0139] The control device 700 of the injection molding machine 10 executes the inference phase. The control device 700 inputs data into the learned model LM, causing the learned model LM to perform inference. Furthermore, the control device 700 sets predetermined molding conditions based on the output from the learned model LM.

[0140] In this embodiment, the structure of the learned model LM generated by the learning unit 1315 of the learning device 1300 and the learned model LM used by the inference unit 715 of the control device 700 can be made universal. In addition, the learning device 1300 can also transmit the weights and deviations of the learned model LM to the control device 700. The control device 700 updates the learned model LM based on the received weights and deviations, thereby making it consistent with the learned model LM of the learning unit 1315 of the learning device 1300. Thus, the learning unit 1315 of the learning device 1300 and the inference unit 715 of the control device 700 of the injection molding machine 10 can be installed in different languages. For example, the learning unit 1315 of the learning device 1300 can be installed using Python (a high-level computer programming language). On the other hand, the inference unit 715 of the control device 700 can be installed using C++, for example.

[0141] Furthermore, the learning device 1300 and the control device 700 can have different CPUs, OSs, development languages, etc. The control device 700 can use a CPU with a lower processing speed than the learning device 1300. This can reduce costs.

[0142] Furthermore, the structural information of the weights and biases transmitted only needs to include the values ​​of each weight, the values ​​of each bias, and parameter structure information. For example, it can also be an object that describes the values ​​of each weight and each bias according to a data structure and through a prescribed method (such as JSON (JavaScript Object Notation: data exchange format), XML (Extensible Markup Language: extensible markup language), ONNX (Open Neural Network Exchange: open neural network exchange), etc.).

[0143] Thus, the learning device 1300 can transmit information indicating the structure of weights and biases to the control device 700 of the injection molding machine 10 .

[0144] Furthermore, this embodiment is not limited to the method of transmitting information indicating the structure of weights and biases from the learning device 1300 to the injection molding machine 10. Alternatively, the learned model LM itself may be transmitted from the learning device 1300 to the control device 700 of the injection molding machine 10. In this case, the CPU, OS, development language, and the like may be made common between the control device 700 and the learning device 1300.

[0145] Thus, the control device 700 can perform inference using the learned model LM generated in the learning phase of the learning device 1300 .

[0146] Figure 4 1 is a diagram showing an example of the functional configuration of the learning device 1300 and the control device 700 according to this embodiment. Figure 4 As shown, the learning device 1300 and the control device 700 are connected via a communication line NW.

[0147] The functions of the learning device 1300 are implemented by any hardware or any combination of hardware and software. Figure 4 As shown, the learning device 1300 includes a CPU 1301 , a storage medium 1302 , and a communication interface 1303 .

[0148] The storage medium 1302 stores various installed programs and files and data required for various processes. The storage medium 1302 includes, for example, a HDD (Hard Disc Drive), an SSD (Solid State Drive), a flash memory, and the like.

[0149] The storage medium 1302 according to this embodiment includes a training data storage unit 1321 and a learned model storage unit 1322 .

[0150] The training data storage unit 1321 stores training data used for learning the learned model LM. The structure of the training data will be described later.

[0151] The learned model storage unit 1322 stores the learned model LM.

[0152] The communication interface 1303 is used as an interface for connecting to external devices for communication. Thus, the learning device 1300 can communicate with external devices such as the injection molding machine 10 through the communication interface 1303. Furthermore, the communication interface 1303 may include multiple communication interfaces depending on the communication method with the connected device.

[0153] The CPU 1301 of the learning device 1300 executes a program stored in the storage medium 1302. Thus, the CPU 1301 includes an acquisition unit 1311, an evaluation acquisition unit 1312, a correct answer information calculation unit 1313, a training data generation unit 1314, a learning unit 1315, and a communication control unit 1316 as functional units.

[0154] The learning device 1300 according to this embodiment generates a learned model LM using the above-described configuration. The learned model LM, for example, outputs information related to the settings of predetermined items of the molding conditions of the injection molding machine 10 when waveform data representing, in a time series, detection results detected by a sensor provided in the injection molding machine 10 during injection molding is input.

[0155] That is, the learned model LM according to the present embodiment outputs information on settings for improving the quality of a molded product when a detection result of a sensor is input.

[0156] Conventionally, when sensor detection results were input into a learned model, there was a tendency to infer product quality regardless of the set values. Consequently, it was necessary to set specific parameters based on the inference results from the previously learned model. This often required multiple steps to complete the setup process.

[0157] In contrast, the injection molding machine 10 according to the present embodiment can be set based on information output from the learned model LM. This improves the quality of the molded product and reduces the processing load.

[0158] Next, a description will be given of a configuration of the learning device 1300 for generating training data used to generate the learned model LM.

[0159] Each time injection molding is performed by setting a setting value in an item to be set, the acquisition unit 1311 acquires waveform data (an example of first data) indicating a detection result detected by a sensor provided in the test injection molding machine 1350 from the test injection molding machine 1350 .

[0160] In this embodiment, the case where the item to be set is a clamping force setting value is described. Furthermore, each time the clamping force setting value is set to mold a molded product, the acquisition unit 1311 acquires waveform data (an example of aggregate data) representing the clamping force detected by a clamping force detector (an example of a detection device) from the start of filling to the end of cooling, for example, from the test injection molding machine 1350. Furthermore, this embodiment does not limit the acquired waveform data to waveform data representing the clamping force from the start of filling to the end of cooling; waveform data representing the clamping force for each injection is sufficient. For example, the acquired waveform data may also represent a portion or the entire molding cycle, or waveform data approximating the previous and subsequent injections.

[0161] The learning device 1300 according to this embodiment generates training data for outputting information related to the setting of a mold clamping force setting value. Various methods are conceivable for generating training data. Furthermore, information indicating correct answers contained in the training data can also be generated using various methods. Therefore, in this embodiment, the first and second methods for generating training data are described.

[0162] (First method for generating training data)

[0163] The first method for generating training data is a method of generating training data based on the presence or absence of burrs on a molded product.

[0164] Each time a clamping force setting value is set and injection molding is performed based on that setting, the evaluation acquisition unit 1312 acquires evaluation information indicating an evaluation of the resulting molded product. For example, the evaluation acquisition unit 1312 acquires information indicating the presence or absence of burrs on the molded product as evaluation information. The method for acquiring the evaluation information may be any method. For example, the evaluation information may be input via an operating device, based on a user's visual inspection of the molded product (not shown), or may be based on detection results from various sensors.

[0165] Figure 5 : is a diagram illustrating evaluation information acquired by the evaluation acquisition unit 1312 according to this embodiment. Figure 5 In the example shown, the presence or absence of burrs on the molded product is displayed in association with each mold clamping force setting value. Figure 5 The presence or absence of burrs on molded products shown is information inputted as a result of visual confirmation by the user.

[0166] according to Figure 5 Based on the correspondence relationship shown, the correct answer information calculation unit 1313 derives the mold clamping force setting value at which burrs are not generated on the molded product as information indicating the correct answer included in the training data.

[0167] In addition, in this embodiment, a method of obtaining the presence or absence of burrs on a molded product when injection molding is performed based on a mold clamping force setting value (an example of the first value) is described each time a mold clamping force setting value is set, but the method is not limited to this method. Figure 5 When the clamping force detection value (an example of the first value) shown in FIG. 1 is used, it is also possible to determine whether the molded product has burrs when the clamping force detection value is detected. The clamping force detection value is, for example, the highest value (peak value) among the clamping force detection values ​​detected from the start to the end of the molding cycle. In addition, the present embodiment is not limited to the peak value and can be an integral value over a specified period (e.g., the molding cycle), a difference (maximum value - minimum value) between detection values ​​detected within a specified period, or a detection value detected at a specific moment depending on the process status, such as the start or end of a specified process.

[0168] Return to Figure 4 The correct answer information calculation unit 1313 generates information related to the correct answer of the clamping force setting value based on the presence or absence of burrs on the molded product (an example of evaluation information) obtained by the evaluation acquisition unit 1312. The presence or absence of burrs on the molded product is evaluation information obtained each time the clamping force setting value is set or each time the clamping force detection value is detected. The correct answer information calculation unit 1313 involved in this embodiment generates information related to the correct answer of the clamping force setting value based on the presence or absence of burrs in multiple shots after obtaining the presence or absence of burrs in multiple shots. That is, in this embodiment, by considering the tendency of the presence or absence of burrs in multiple shots, it is possible to accurately generate information related to the correct answer of the clamping force setting value. In this embodiment, a recommended clamping force value or a clamping force adjustment value is generated as information related to the correct answer of the clamping force setting value. In addition, in this embodiment, the case where the evaluation information is the presence or absence of burrs is described, but the evaluation information is not limited to the presence or absence of burrs, as long as it is information that serves as a benchmark for the quality of the molded product. For example, the evaluation information may include the weight of the molded product, the area of ​​a predetermined surface of the molded product, the length of a predetermined contour of the molded product, and the like.

[0169] Figure 6 FIG. 1 is a diagram showing an example of a method for generating information indicating a correct answer to a mold clamping force setting value by the correct answer information calculation unit 1313 according to this embodiment. Figure 6In the example shown, burrs disappear when the clamping force setting is 700 kN or greater. Therefore, the correct answer information calculation unit 1313 derives a recommended clamping force value of 700 kN as information indicating a correct answer. The recommended clamping force value is the recommended setting for clamping force to suppress burrs.

[0170] In addition, while the correct answer information calculation unit 1313 of this embodiment derives the minimum value among the clamping force setting values ​​that do not produce burrs as the recommended clamping force value, the correct answer information calculation unit 1313 is not limited to this derivation method. For example, the correct answer information calculation unit 1313 may also derive the intermediate value (e.g., 650 [kN]) between the minimum value among the clamping force setting values ​​that do not produce burrs and the maximum value among the clamping force setting values ​​that produce burrs as the recommended clamping force value.

[0171] The correct answer information calculation unit 1313 of this embodiment generates information related to the correct answer for each clamping force setting value. As described above, the information related to the correct answer can be a recommended clamping force value or a clamping force adjustment value indicating an adjustment amount for the currently set clamping force setting value. Alternatively, the correct answer information calculation unit 1313 can generate information related to the correct answer for each clamping force detection value, rather than for each clamping force setting value. In this case, the correct answer information calculation unit 1313 can calculate the adjustment amount for the detected clamping force value as the information related to the correct answer.

[0172] Figure 7 This is a diagram illustrating information related to the correct answer generated by the correct answer information calculation unit 1313 according to this embodiment. Figure 7 As shown, the information related to the correct answer generated for each mold clamping force setting value may be a recommended mold clamping force value or an adjusted mold clamping force value. Figure 7 The recommended clamping force value or the clamping force adjustment value shown is included in the training data as information indicating the correct answer.

[0173] Return to Figure 4 The training data generation unit 1314 generates training data by combining the waveform data for each clamping force setting value (or clamping force detection value) acquired by the acquisition unit 1311 and the recommended clamping force value or clamping force adjustment value for each clamping force setting value (or clamping force detection value) generated by the correct answer information calculation unit 1313. The training data generation unit 1314 stores the generated training data in the training data storage unit 1321.

[0174] (Second method for generating training data)

[0175] The second method for generating training data is a method of generating training data based on measurement results corresponding to the amount of burrs.

[0176] Each time a mold clamping force setting value is set and injection molding is performed based on that setting, the evaluation acquisition unit 1312 acquires evaluation information indicating an evaluation of the molded product. For example, the evaluation acquisition unit 1312 acquires a value indicating a measurement result (hereinafter referred to as a measurement value) obtained by a sensor installed in the injection molding machine 10 as the evaluation information.

[0177] The measured value may be, for example, the difference in mold clamping force generated during the molding cycle (detected by a mold clamping force detector). Figure 8 This is a diagram illustrating the difference in mold clamping force acquired by the evaluation acquisition unit 1312 according to this embodiment. Figure 8 The line 1801 shown represents the change in the mold clamping force from the start of filling to cooling in any one shot.

[0178] Furthermore, the evaluation acquisition unit 1312 acquires a value obtained by subtracting the mold clamping force detection value P11 detected at the filling start time t11 from the mold clamping force detection value P12 detected at the time t12 before the mold device 800 is opened as the mold clamping force difference. Figure 8 In the example shown, the difference in clamping force (in other words, the increased clamping force) is considered to be the amount by which the mold assembly 800 is opened due to the molding material entering the parting line surface. Furthermore, the evaluation acquisition unit 1312 calculates the difference in clamping force and acquires the calculated difference in clamping force (clamping force P12 - clamping force P11) as a measured value corresponding to the thickness of the burrs generated on the molded product. In the description of this generation method, the burr thickness is used as an example of the burr amount. Furthermore, in this generation method, the difference in clamping force corresponding to the burr thickness is used as an example of the measurement result corresponding to the burr amount.

[0179] In addition, the present generation method does not limit the measured value corresponding to the thickness of the burrs generated on the molded product to the value obtained by subtracting the detected value of the mold clamping force P11 from the detected value of the mold clamping force P12. For example, as the difference in the mold clamping force, the difference between the peak value of the mold clamping force and the mold clamping force detected at the start of filling can also be used. The amount of burrs involved in the present generation method is not limited to the thickness of the burrs, and the height or length of the burrs can also be considered. That is, compared with the presence or absence of burrs shown in the first generation method mentioned above, the amount of burrs involved in the present generation method can be a continuous value obtained by calculation or measurement. In addition, as the measured value corresponding to the amount of burrs, for example, a measured value corresponding to any one or more of the thickness, height and length of the burrs can also be used. There are various types of measured values ​​corresponding to the amount of burrs. In the present generation method, any of these various types of measured values ​​can be used as the measured value corresponding to the amount of burrs.

[0180] Figure 9: is a diagram illustrating the difference in mold clamping force acquired by the evaluation acquisition unit 1312 according to the modification. Figure 9 In the example shown, line 1901 represents the detection value of the mold clamping force detected by the mold clamping force detector. Figure 9 The line 1901 shown represents the change in the mold clamping force from the start of filling to cooling in any one shot.

[0181] Furthermore, the evaluation acquisition unit 1312 acquires the value obtained by subtracting the mold clamping force detection value P21 detected at the start of filling t21 from the mold clamping force detection value P22 detected at the time t22 when the mold clamping force is maximum, as the mold clamping force difference. In this embodiment, the mold clamping force difference (in other words, the increase in mold clamping force) is considered to be the amount by which the mold assembly 800 is opened due to the filling of the molding material during the filling / holding pressure process. Furthermore, the evaluation acquisition unit 1312 calculates the mold clamping force difference and acquires this mold clamping force difference (the mold clamping force detection value P22 minus the mold clamping force detection value P21) as a measured value corresponding to the thickness of the burrs generated in the molded product (an example of evaluation information).

[0182] Figure 10 : is a diagram illustrating the measurement values ​​acquired by the evaluation acquisition unit 1312 according to this embodiment. Figure 10 In the example shown, the difference in mold clamping force is associated with each mold clamping force setting value as a measured value. Figure 10 The measured value shown is a value obtained by subtracting the mold clamping force detection value detected at the start of filling from the mold clamping force detection value detected at a time before the mold device 800 is opened.

[0183] according to Figure 10 Based on the correspondence relationship shown, the correct answer information calculation unit 1313 calculates the mold clamping force setting value at which burrs are not generated on the molded product as information indicating the correct answer.

[0184] In addition, in this embodiment, the example of obtaining the measured value when the mold clamping force setting value is set (indicating the evaluation of the presence or absence of burrs) is described each time the mold clamping force setting value (an example of the first value) is set. However, the method is not limited to obtaining the measured value for each mold clamping force setting value. For example, each time a mold clamping force setting value is detected, Figure 10 When the mold clamping force detection value (an example of the first value) is shown, a measurement value (indicating an evaluation of the presence or absence of burrs) is acquired when the mold clamping force detection value is detected.

[0185] The correct answer information calculation unit 1313 generates information on the correct answer of the mold clamping force setting value based on the acquired measurement value (an example of evaluation information) each time the mold clamping force setting value is set or the mold clamping force detection value is detected.

[0186] Figure 111 is a diagram showing an example of a method for calculating information related to a correct answer of a mold clamping force setting value by the correct answer information calculation unit 1313 according to this embodiment. Figure 11 In the example shown, the correct information calculation unit 1313 approximates the measured value of each mold clamping force using two straight lines, line 2101 and line 2102. As an approximation method for the two straight lines, for example, the least squares method is used. Then, the correct information calculation unit 1313 calculates the intersection of the two straight lines as the recommended mold clamping force value. Figure 11 In the example shown, the correct answer information calculation unit 1313 calculates the mold clamping force 680 [kN], which is the intersection of the line 2101 and the line 2102, as the recommended mold clamping force value. Figure 11 This is just one example of a calculation method, and the method is not limited to this one. For example, the correct answer information calculation unit 1313 may perform curve approximation instead of two-line approximation, and calculate the change point of the curve as the recommended mold clamping force value.

[0187] The correct answer information calculation unit 1313 of this embodiment calculates information related to the correct answer each time a clamping force setting value is set or each time a clamping force detection value is detected. As described above, the correct answer information can be a recommended clamping force value or a clamping force adjustment value indicating an adjustment amount for the currently set clamping force setting value. Alternatively, the information can be an adjustment amount for the currently detected clamping force detection value. In the case of an adjustment amount for the clamping force detection value, the detected clamping force detection value is adjusted by the adjustment amount and set as the clamping force setting value.

[0188] Figure 12 : is a diagram illustrating information related to the correct answer calculated by the correct answer information calculation unit 1313 involved in this embodiment. Figure 12 As shown, the information related to the correct answer calculated for each mold clamping force setting value may be a recommended mold clamping force value or an adjusted mold clamping force value. Figure 12 The recommended clamping force values ​​or clamping force adjustment values ​​shown are used as training data as information indicating correct answers.

[0189] Return to Figure 4 The training data generation unit 1314 generates training data by combining the waveform data for each clamping force setting value (or clamping force detection value) acquired by the acquisition unit 1311 and the recommended clamping force value or clamping force adjustment value for each clamping force setting value (or clamping force detection value) generated by the correct answer information calculation unit 1313. The training data generation unit 1314 stores the generated training data in the training data storage unit 1321.

[0190] This embodiment illustrates a method for generating training data, but training data may be generated by methods other than the above method. In addition, in this embodiment, an example is described in which the correct answer information calculation unit 1313 automatically generates information related to the correct answer based on the evaluation information. However, the method is not limited to automatically generating information based on the evaluation information. For example, the correct answer information calculation unit 1313 may also determine information related to the correct answer based on information input by the user via the operating device. As a specific example, Figure 11 As shown, the correct answer information calculation unit 1313 approximates each measured value of the mold clamping force using two straight lines and outputs the approximated result to the display device. The user refers to the approximated result displayed on the display device and inputs information related to the correct answer, such as a recommended mold clamping force value, via the operating device. Furthermore, the correct answer information calculation unit 1313 determines information related to the correct answer based on the information input via the operating device. Furthermore, in this embodiment, after calculating the information related to the correct answer, the correct answer information calculation unit 1313 may also accept visual confirmation of the correct answer information by the user and may modify the information.

[0191] Furthermore, the method is not limited to one in which the learning device 1300 automatically performs all steps until the generation of the learning model LM is completed. For example, after the learning device 1300 generates a plurality of training data, the user may confirm the contents of each of the plurality of training data.

[0192] In this embodiment, the plurality of training data may be referred to as a training data set. The training data set is a collection of training data based on a combination of waveform data as input and recommended mold clamping force values ​​or adjusted mold clamping force values ​​as output.

[0193] Then, after the user completes confirmation and correction of the contents included in the training data set, the learning device 1300 can generate a learned model LM based on the training data set.

[0194] (Generation of learning-completed models, etc.)

[0195] The learning unit 1315 generates a learned model LM by performing machine learning based on the training data set stored in the training data storage unit 1321. The learned model LM is generated by applying supervised learning to a basic learning model.

[0196] Specifically, the learning unit 1315 performs machine learning based on training data composed of waveform data representing the results detected by a clamping force detector during molded product production for each clamping force setting value (for example, a first value) set for producing a molded product by the injection molding machine 10, and recommended clamping force values ​​or adjusted clamping force values ​​representing the correct answer regarding the setting of the clamping force setting value, derived from evaluation information (the presence or absence of burrs or measured values) indicating an evaluation of the molded product produced with that value set for each clamping force setting value (for example, a first value). A learned model LM is then generated as a result of this machine learning.

[0197] The learned model LM is a model that outputs a recommended mold clamping force value or a mold clamping force adjustment value (an example of information related to the mold clamping force setting value of the molding condition) from the output layer when waveform data (an example of the second data) representing the timing of the mold clamping force detected during molding in the injection molding machine 10 or the test injection molding machine 1350 is input from the input layer.

[0198] The waveform data input from the input layer of the learned model LM is aggregate data representing, in time series, the results detected by the mold clamping force detector provided in the injection molding machine 10 while the injection molding machine 10 is producing a molded product.

[0199] The information output from the output layer of the learned model LM can be a recommended mold clamping force value (an example of a setting value) that should be set as the mold clamping force setting value (an example of a prescribed item), or a mold clamping force adjustment value (an example of a correction value) for the currently set mold clamping force setting value.

[0200] Furthermore, the learned model LM may be updated by additionally learning a new training data set on the existing learned model LM.

[0201] As machine learning for generating the learned model LM, for example, a neural network can be applied. As another example, deep learning, which is machine learning using a deep neural network (DNN), can also be applied. Deep learning, for example, can also be applied using convolutional neural networks, RNNs (recurrent neural networks), and LSTMs (long short-term memories).

[0202] Figure 13 This is a diagram illustrating the configuration of a learned model LM used in the learning device 1300 and the control device 700 according to the first embodiment.

[0203] Figure 13 The learned model LM shown is composed of an input layer 2311, one or more intermediate layers 2312 connected to the input layer, and an output layer 2313 connected to the intermediate layers. The input layer has, for example, 2200 nodes, the intermediate layer has, for example, 200 nodes, and the output layer has 1 node.

[0204] The learned model LM outputs output data 2302 corresponding to one node when input data 2301 corresponding to 2200 nodes is input.

[0205] The input data 2301 is, for example, waveform data that chronologically represents actual values ​​that change during injection molding. The input data 2301 may also be waveform data that chronologically represents changes in the mold clamping force, which is the detection result of a mold clamping force detector, during injection molding.

[0206] The output data 2302 outputs, for example, parameters for adjusting predetermined items of the injection molding machine 10. The output data 2302 may be, for example, a clamping force recommendation value or a clamping force adjustment value for adjusting a clamping force setting value set by the injection molding machine 10 during injection molding.

[0207] That is, the learned model LM outputs a recommended mold clamping force value recommended to be set as the mold clamping force setting value in the injection molding machine 10, or a mold clamping force adjustment value for adjusting the current mold clamping force setting value, when waveform data representing the time-series changes in the mold clamping force detected by the mold clamping force detector is input.

[0208] When the learning unit 1315 performs learning using the training data, the waveform data included in the training data may be pre-processed. In addition, the pre-processing may also be performed when the inference unit 715 described later performs inference.

[0209] In this embodiment, for example, preprocessing can be performed as follows: the clamping force setting value or detection result used in the waveform data is divided by the rated clamping force so that the clamping force setting value, detection result, recommended clamping force value, etc. converge within the range of 1.0.

[0210] In the test injection molding machine 1350 of this embodiment, various mold devices are installed and various molding conditions are set. In addition, the learning device 1300 generates a training data set based on various inspection results of molded products produced by the test injection molding machine 1350.

[0211] In this manner, the learning device 1300 prepares a training data set based on the various molding conditions and mold assemblies set in the test injection molding machine 1350. The learning device 1300 performs machine learning using this training data set to generate a learned model LM. This allows inference using the learned model LM to be performed regardless of the molding conditions or the type of mold assembly 800.

[0212] Thus, in the control device 700 of the injection molding machine 10 , regardless of the installed mold device 800 and the set molding conditions, the settings can be changed using the learned model LM.

[0213] exist Figure 13 In the example shown, in the input layer, the variables corresponding to each node are defined as the input variable vector x 0～2199 In the middle layer, the variables corresponding to each node are defined as the intermediate variable vector z 0～199 In the output layer, the variable corresponding to the node is defined as the output variable y.

[0214] At this time, the formula for calculating the intermediate variable vector z0 of the intermediate layer is calculated by the following formula (1). In formula (1), the weight vector w 0～199,0～2199 Set as the intermediate variable vector z output from each node of the input layer 0～199 The weight vector (200×2200 dimension) of each value of . In addition, the intermediate variable vector z 1～199 It can also be calculated by using the same formula, so the explanation is omitted. Figure 13 The deviation b is omitted in 0～199 , but it is set as the bias used in the neural network. The activation function φ can be considered to use the ReLU function, for example, but other functions can also be used.

[0215] z0=φ(w 0,0～2199 ·x 0～2199 +b0)……(1)

[0216] The formula for calculating the output variable y of the output layer is calculated by the following formula (2). In formula (2), the weight vector w' 0～199 Let be the weight vector used to output values ​​from each node in the middle layer to the output variable y. Figure 13 The bias b' is omitted in , but it is set as the bias used in the neural network. The activation function φ' can be considered to use the ReLU function, for example, but other functions can also be used.

[0217] y=φ'(w' 0～199 ·z 0～199 +b')……(2)

[0218] The learning unit 1315 stores the generated learned model LM in the learned model storage unit 1322 .

[0219] Return to Figure 4 The communication control unit 1316 uses the communication I / F 1303 to transmit and receive information to and from external devices such as the injection molding machine 10. For example, the communication control unit 1316 may transmit the learned model LM stored in the learned model storage unit 1322 to the control device 700 of the injection molding machine 10. In addition, the communication control unit 1316 may extract a structure representing parameters (e.g., weights and biases) set for each layer constituting the learned model LM stored in the learned model storage unit 1322 and transmit the structure to the control device 700.

[0220] exist Figure 4 Components of the control device 700 of the injection molding machine 10 are shown as functional blocks. Figure 4 The functional blocks shown in the figure are conceptual functional blocks and do not necessarily need to be physically constructed as shown. All or part of each functional block can be functionally or physically dispersed / integrated in arbitrary units to form a structure. All or any part of each processing function performed in each functional block is implemented by a program executed by the CPU 701. Alternatively, each functional block can also be implemented as hardware based on wired logic. Figure 4 As shown, the CPU 701 of the control device 700 includes a communication control unit 711, an update unit 712, an injection molding processing unit 713, and a setting unit 714. Furthermore, the control device 700 includes a learned model storage unit 721 in the storage medium 702.

[0221] The learned model storage unit 721 stores the learned model LM. The structure of each layer of the learned model LM is the same as that of the learned model LM stored in the learned model storage unit 1322 of the learning device 1300.

[0222] The communication control unit 711 uses the communication I / F 705 to transmit and receive information to and from external devices such as the learning device 1300. For example, the communication control unit 711 may receive the learned model LM from the learning device 1300. Furthermore, the communication control unit 711 may receive information representing a structure of parameters (e.g., weights and biases) set in each layer constituting the learned model LM.

[0223] In addition, in this embodiment, the method of acquiring the learned model LM or the information representing the structure is described using an example of receiving the information from the learning device 1300. However, the method of acquiring the learned model LM or the information representing the structure is not limited thereto. For example, the learned model LM or the information representing the structure may also be acquired via an external storage medium.

[0224] The updating unit 712 updates the learned model LM stored in the learned model storage unit 721 using the received learned model LM or information indicating the structure. This allows the updating unit 712 to make the learned model LM stored in the learned model storage unit 721 consistent with the learned model LM of the learning device 1300.

[0225] In the case of updating using information representing a structure representing parameters (e.g., weights and biases) set in each layer constituting the learned model LM, the update unit 712 may be configured as a program for updating the learned model LM. The update unit 712 may also determine whether the number of layers and the number of nodes per layer defined between the learned model LM stored in the learned model storage unit 721 and the received information representing the structure are consistent. Furthermore, if the update unit 712 determines that they are consistent, it updates the learned model LM. If it determines that they are inconsistent, the update unit 712 may not update the learned model LM and instead output an alarm screen with a different structure to the display device 760, etc.

[0226] The injection molding processing unit 713 executes processing for producing a molded product using the injection molding machine 10. For example, when producing a molded product, the injection molding processing unit 713 may perform injection molding after setting various items constituting molding conditions.

[0227] The setting unit 714 includes an inference unit 715 and sets a mold clamping force setting value based on information (for example, a recommended mold clamping force value or an adjusted mold clamping force value) acquired from the inference unit 715 .

[0228] The inference unit 715, for example, inputs waveform data (an example of the second data) representing the results detected by the clamping force detector during the production of the molded product by the injection molding processing unit 713 in a time series into the input layer of the learned model LM, and receives a recommended clamping force value or a clamping force adjustment value from the output layer of the learned model LM.

[0229] Furthermore, the inference unit 715 may pre-process the waveform data before inputting it into the input layer.

[0230] The setting unit 714 then sets the clamping force set value based on the recommended clamping force value or the adjusted clamping force value received from the inference unit 715. The setting unit 714 of this embodiment may also set a limit on the recommended clamping force value or the adjusted clamping force value output from the learned model LM when setting the clamping force set value, so that the value is set within a range of a predetermined upper limit (or lower limit). Furthermore, the setting unit 714 of this embodiment may set the clamping force set value by multiplying the recommended clamping force value or the adjusted clamping force value output from the learned model LM by a predetermined safety factor or by adding a predetermined margin.

[0231] While this embodiment describes an example in which the learned model LM outputs recommended clamping force values ​​or adjusted clamping force values, the information output by the learned model LM is not limited thereto. Specifically, the information output by the learned model LM can be any information related to items that can be set as molding conditions. For example, in addition to recommended and adjusted values, the information output by the learned model LM can also include detected values ​​related to the molded product, such as those from which recommended and adjusted values ​​can be derived for specified items.

[0232] Conventionally, when detection results from sensors installed in injection molding machines are fed into the input layer of a learned model, information related to the evaluation of molded products produced by the injection molding machine is often output. Previously, it was possible to determine whether a molded product was defective, for example, based on the evaluation output from the learned model. However, in order to derive settings to prevent defective molded products, the information output from the learned model required further transformation, taking into account the relationship between the degree of defectivity and the set values.

[0233] In contrast, in this embodiment, the clamping force setting value is set using the recommended clamping force value or the clamping force adjustment value output from the learned model LM. Specifically, in this embodiment, the injection molding machine 10 can achieve improved quality accuracy of molded products produced after the setting. Furthermore, the clamping force setting value can be automatically set, thereby reducing the operational burden on the user and the processing burden on the control device 700.

[0234] In the inference unit 715 involved in this embodiment, each time waveform data (an example of the second data) is input to the input layer of the learned model LM, a recommended mold clamping force value or a mold clamping force adjustment value is received from the output layer. Moreover, the setting unit 714 sets the mold clamping force setting value based on the recommended mold clamping force value or the mold clamping force adjustment value received from the inference unit 715. That is, the control device 700 involved in this embodiment can set the mold clamping force setting value in units of one injection. The control device 700 involved in this embodiment can quickly set the mold clamping force setting value, so it can start work quickly and achieve improved work efficiency. Regarding the following modified examples and implementations, it can also be set at each injection, so it is the same as the present embodiment in that it can achieve improved work efficiency.

[0235] (Variation 1)

[0236] In the above embodiment, an example is described in which waveform data representing detection results from a sensor (e.g., a mold clamping force detector) installed in the injection molding machine 10 is input into the input layer of the learned model LM. However, the above embodiment is not limited to the method of inputting waveform data into the input layer of the learned model LM. Therefore, in Modification 1, another method of inputting data into the input layer is described.

[0237] The acquisition unit 1311 according to this variation acquires aggregate data representing measurement results of a molded product after the molded product is produced from the test injection molding machine 1350 for each mold clamping force setting value set for molding the molded product. Specifically, the acquisition unit 1311 may acquire, as aggregate data, the amount of burrs measured at a plurality of predetermined measurement locations on the molded product (an example of measurement results).

[0238] The evaluation acquisition unit 1312 and the correct answer information calculation unit 1313 perform the same processing as in the above-described embodiment, and therefore description thereof will be omitted.

[0239] Moreover, the training data generating unit 1314 generates training data by combining the aggregate data representing the amount of burrs respectively measured by multiple measurement values ​​of the molded product acquired by the acquiring unit 1311 and the recommended mold clamping force value or the mold clamping force adjustment value for each mold clamping force setting value generated by the correct information calculating unit 1313.

[0240] In this way, the learning device 1300 can generate a learned model LM based on the training data.

[0241] After a molded article is produced by the injection molding machine 10, the control device 700 of the injection molding machine 10 receives input as aggregate data, including the amount of burrs measured at a plurality of pre-set measurement locations on the molded article (an example of measurement results). The control device 700 then inputs the aggregate data into the input layer of the learned model LM, thereby receiving a recommended mold clamping force value or an adjusted mold clamping force value from the output layer, and sets the mold clamping force set value based on the received recommended mold clamping force value or adjusted mold clamping force value.

[0242] This modification shows an example of the aggregate data used as training data in learning device 1300 and the aggregate data input to the input layer of control device 700. However, the aggregate data is not limited to the burr amounts measured at multiple measurement locations on a molded product. In other words, any aggregate data representing measurement results for a molded product after production will suffice.

[0243] For example, the aggregate data used as training data in the learning device 1300 and the aggregate data input to the input layer in the control device 700 may be image data obtained by photographing produced molded products. Furthermore, a combination of these aggregate data may be used.

[0244] That is, in this modification and the above-described embodiment, it is sufficient to generate and infer training data using aggregated data representing results detected during production of molded products or results measured after production of molded products.

[0245] (Variation 2)

[0246] The above embodiments and variations describe examples of setting the mold clamping force setting value. However, the above embodiments and variations are not limited to setting the mold clamping force setting value. Therefore, in Variation 2, the setting of the dwell time is described.

[0247] The test injection molding machine 1350 according to the present embodiment molds a molded product at predetermined time intervals for a holding pressure time of 1.0 to 8.5.

[0248] The acquisition unit 1311 according to this modification acquires waveform data (an example of first data) that displays, in time series, the detection results of the sensor installed in the test injection molding machine 1350 for each dwell time. In this modification, waveform data that displays, in time series, one or more of the position of the screw 330, the holding pressure, and the mold clamping force, for example, is acquired as the detection result.

[0249] The evaluation acquisition unit 1312 acquires, for each dwell time, evaluation information indicating the evaluation of the molded product produced under the specified dwell time. For example, the evaluation acquisition unit 1312 acquires the weight of the molded product for each mold clamping force setting as evaluation information. The method for acquiring the weight of the molded product may be any method. For example, the weight of the molded product removed from the injection molding machine 10 may be automatically measured, or the user may manually measure the weight.

[0250] Figure 14 : is a diagram illustrating evaluation information acquired by the evaluation acquisition unit 1312 according to this modification. Figure 14 In the example shown, the weight of the molded product is shown in association with each maintenance time.

[0251] according to Figure 14 Based on the correspondence relationship shown, the correct answer information calculation unit 1313 derives the accurate dwell time for molding the molded product.

[0252] Figure 15 : is a graph showing the corresponding relationship between the holding time and weight involved in this modification. Figure 15 In the example shown, the horizontal axis is the holding time and the vertical axis is the weight. Figure 15 The corresponding relationship between the weight and the holding time can be used to derive the gate sealing time. Figure 15 As shown by line 2501, as the dwell time increases, more molding material is fed into the cavity space 801, causing the weight to increase. However, after a certain period of time, the weight stops changing. The time when the weight stops changing is the gate sealing time.

[0253] That is, the correct answer information calculation unit 1313 refers to Figure 15 Line 2501 derives the gate sealing time, and calculates the recommended value of the holding time or the adjusted value of the holding time based on the gate sealing time.

[0254] For example, the correct answer information calculation unit 1313 refers to Figure 15 Line 2501 indicates that the time 7.5, when the weight no longer changes, is determined as the gate sealing time, and the time 8.25, obtained by multiplying the gate sealing time by 1.1, is set as the recommended holding time value. Furthermore, the correct answer information calculation unit 1313 may also use the time obtained by subtracting the current holding time from the multiplied time as the holding time adjustment value.

[0255] Figure 16 This is a diagram illustrating information related to the correct answer generated by the correct answer information calculation unit 1313 according to this modification. Figure 16 As shown, the information related to the correct answer generated based on each holding time can be a recommended holding time value or an adjusted holding time value. Figure 16The recommended dwell time values ​​or adjusted dwell time values ​​shown are used as training data as data that will become correct answers.

[0256] The following processing is the same as that in the above-described embodiment, and therefore its description is omitted. In the above-described embodiment and variations, examples are described for setting the clamping force setting value or the holding pressure time. However, the above-described embodiment and variations do not limit the setting objects to the clamping force setting value or the holding pressure time. The setting objects may be any information used for injection molding, particularly information that needs to be adjusted according to the conditions and environment in which the injection molding is performed. Setting objects may also apply to, for example, VP switching position, holding pressure setting, filling speed setting, filling pressure, and back pressure setting. Furthermore, metering-related setting information may also apply to metering speed, metering delay, and metering end position. Furthermore, in addition to the clamping force setting value, clamping-related setting information may also apply to pressure rise timing, mold opening and closing speed, and mold opening position. Furthermore, ejection position, ejection pressure, ejection speed, and ejection compression timing may also be applied as setting information related to the ejector device 200. Furthermore, cylinder temperature setting, nozzle temperature setting, water cooling temperature, and mold temperature may also be applied as settings related to the temperature within the injection molding machine 10.

[0257] (Second embodiment)

[0258] In the above embodiment, an example is described in which the control device 700 of the injection molding machine 10, as the management device of the injection molding machine 10, uses a learned model to perform settings. However, the above embodiment is not limited to the method in which the control device 700 of the injection molding machine 10 uses a learned model to perform settings. Therefore, in the second embodiment, an example is described in which the group management device 2700 that controls the injection molding machine 10 uses a learned model LM to perform settings.

[0259] Figure 17 1 is a diagram illustrating the configuration of the learning device 1300, the group management device 2700, and the injection molding machine 10 according to this embodiment. Figure 17 As shown, the group management device 2700 manages, for example, eight injection molding machines 10. The number of injection molding machines to be managed is just an example, and any number may be used.

[0260] In this embodiment, the setting using the learned model LM by the control device 700 shown in the above embodiment is applied to a group management device 2700 having a group management function for a plurality of injection molding machines 10 .

[0261] The learning device 1300 according to this embodiment has the same configuration as that of Embodiment 1. The learning device 1300 transmits information on the learned model LM to the group management device 2700 via the communication line NW.

[0262] The communication line NW is, for example, an Internet communication line. When communicating between the learning device 1300 and the group management device 2700, it is preferable to connect via a VPN (Virtual Private Network). Connecting via a VPN can improve the security of communication.

[0263] The group management device (an example of a management device) 2700 is a device that manages multiple injection molding machines 10 in terms of productivity, and is connected to each injection molding machine 10. Like the above-mentioned control device 700, it receives molding conditions and detection results obtained from various sensors, and assists in the management and planning of production conditions.

[0264] The group management device 2700 may be realized by a personal computer, for example. However, the group management device 2700 generally does not have a control function for the injection molding operation of each injection molding machine 10. However, if the function of the personal computer is expanded, it can also have a control function.

[0265] Similar to the control device 700 , the group management device 2700 includes a storage device (not shown), and the learned model LM is stored in the storage device.

[0266] Similar to the control device 700, the group management device 2700 includes a CPU (not shown), and by having the CPU execute a program stored in a storage device, a communication control unit 711, an update unit 712, an injection molding processing unit 713, and a setting unit 714 including an inference unit 715 are implemented, similar to the control device 700. The processing executed by each component is the same as in the above embodiment, and therefore, description thereof is omitted.

[0267] The group management device 2700 is connectable to the learning device 1300 via a communication line NW.

[0268] The group management device 2700 receives information related to the learned model LM from the learning device 1300. The group management device 2700 updates the learned model LM based on the received information.

[0269] As an auxiliary management and planning of production conditions, for example, the group management device 2700 can also calculate recommended values ​​or adjustment values ​​for specified items based on the waveform data shown in the received detection results, and send a request to each injection molding machine 10 to set the recommended values ​​or adjustment values.

[0270] <Function>

[0271] In the above embodiment, by using the learned model generated by the learning device 1300, the injection molding machine 10 can be configured for injection molding based on sensor detection results during molding of the molded product or measurement results after the molded product is formed. This improves the accuracy of the molded product. Furthermore, since settings can be made based on information output from the learned model LM, the configuration burden can be reduced.

[0272] Specifically, in the above-described embodiment, after preprocessing data obtained from sensors of the injection molding machine 10 or external measuring devices (including cameras), the learned model LM is used to generate recommended or adjusted values ​​in an end-to-end manner. Therefore, compared to conventional methods in which the molded product is evaluated and then recommended or adjusted values ​​are generated based on that evaluation, the processing is simplified. In the above-described embodiment and its variations, the processing burden can be reduced compared to conventional methods. Consequently, the use of a high-performance CPU, etc., is eliminated, thereby achieving cost reductions.

[0273] While the embodiments of the learned model, injection molding machine management device, and training data generation method according to the present invention have been described above, the present invention is not limited to the aforementioned embodiments. Various changes, modifications, substitutions, additions, deletions, and combinations are possible within the scope of the technical proposals. These are naturally within the technical scope of the present invention.

Claims

1. A learned model comprising an input layer, one or more intermediate layers connected to the input layer, and an output layer connected to the intermediate layer, wherein: The learned model is used to enable the computer to function in the following manner: performing machine learning based on first data and correct answer information, wherein the first data represents, for each first value set or detected in a prescribed item for producing a molded product by an injection molding machine, a result detected during production of the molded product or a result of measuring the molded product after production of the molded product, and the correct answer information represents, as a correct answer, information related to the setting of the prescribed item derived from evaluation information representing an evaluation of the molded product produced when the first value was set or when the first value was detected for each first value, When second data indicating a result detected during production or a result measured after production is input from the input layer, information related to the setting of the predetermined item is output from the output layer.

2. The learning completion model according to claim 1, wherein: The information related to the setting outputted from the output layer is a setting value for the predetermined item or a correction value for a value currently set in the predetermined item.

3. The learning completion model according to claim 1, wherein: The first data and the second data are aggregate data that represent, in time series, results detected by a detection device provided in the injection molding machine during production of the molded product.

4. The learning completion model according to claim 1, wherein: The first data and the second data are aggregate data indicating measurement results of the molded product after the molded product is produced.

5. The learned model according to any one of claims 1 to 4, wherein: The predetermined item is a mold clamping force set in the injection molding machine or a holding pressure time set in the injection molding machine.

6. A management device for an injection molding machine, comprising: A learning-completed model having an input layer, one or more intermediate layers connected to the input layer, and an output layer connected to the intermediate layer, performs machine learning based on first data and correct answer information, wherein the first data represents, for each first value set or detected in a prescribed item for producing a molded product by an injection molding machine, a result detected during the production of the molded product or a result of measuring the molded product after the production of the molded product, and the correct answer information represents, as a correct answer, information related to the setting of the prescribed item derived from evaluation information representing an evaluation of the molded product produced when the first value is set or when the first value is detected for each first value; and The inference unit receives second data from the input layer and acquires information related to the setting of the predetermined item from the output layer, wherein the second data indicates a result detected during production or a result measured after production.

7. The management device for an injection molding machine according to claim 6, wherein: During a cycle in which the injection molding machine produces the molded product, the inference unit inputs the second data from the input layer and acquires information related to the setting of the predetermined item from the output layer.

8. The management device for an injection molding machine according to claim 6, further comprising: The setting unit sets the predetermined item based on the information acquired by the inference unit.

9. A method for generating training data, which is a method for generating training data for machine learning, wherein: acquiring first data representing, for each first value set or detected as a prescribed item for producing a molded product by an injection molding machine, a result detected during production of the molded product or a result measured after production of the molded product, acquiring evaluation information indicating, for each of the first values, an evaluation of the molded product produced when the first value is set or when the first value is detected, generating information related to a correct answer to the predetermined item based on the evaluation information acquired for each of the first values; Training data is generated by combining information related to a correct answer to the predetermined item and the first data.

Images