Manufacturing device for resin molded products, learning device and inference device

Abstract

To provide a technique that accurately detects the occurrence of burrs, independent of the die shape.SOLUTION: A device for producing a resin molded article is designed to accurately detect the occurrence of burrs in a resin product, independent of the shapes of a fixing side die 105 and a movable side die 104. The device for producing a resin molded article includes the fixing side die 105, the movable side die 104, a first sensor 206 provided near the gate of a cavity 205, formed by the fixing side die 105 and the movable side die 104, and a determination unit 321 that detects the occurrence of burrs on the basis of a clearance between the fixing side die 105 and the movable side die 104 by the first sensor 206.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to a manufacturing apparatus for a resin molded product, and more particularly to a technique for detecting burrs on a resin molded product. [Background technology]

[0002] Injection molding is a known method for producing resin molded products (hereinafter sometimes referred to as "molded products"), in which heated and molten resin is injected into a mold and cooled to produce a molded product. During the resin molding process, injection molds are subjected to external forces from machines such as motors and the resin itself. If the mold cannot withstand the external forces, gaps form at the parting surfaces, and resin leaks through these gaps. This results in burrs forming on the molded product. Therefore, molded product manufacturing sites require inspection and deburring processes to check for the presence of burrs on the manufactured molded products, and these processes reduce the productivity of molded products. Therefore, a technology is needed to accurately determine the occurrence of burrs on molded products.

[0003] Regarding technology for detecting burrs in molded products, for example, Japanese Patent Application Laid-Open No. 2008-249612 (Patent Document 1) discloses a burr detection method that "detects burrs that occur when a material is supplied into a molding die in which a movable die is closely attached to a fixed die and held with a predetermined clamping force, and pressure-molded, by measuring the displacement of the outer surface of the molding die during pressure-molding of the material supplied into the molding die, and determining that a burr has occurred when the displacement exceeds a preset reference value" (see [Abstract]). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-249612 Summary of the Invention [Problem to be solved by the invention]

[0005] As a result of further investigations, the inventors discovered that when a mold is filled with resin and subjected to pressure from the inside, it displaces (or curves) in the mold opening direction. More specifically, they discovered that deflection of the mold occurs starting from the mold gate. The technology disclosed in Patent Document 1 does not take into consideration the fact that mold deflection does not occur uniformly throughout the mold but occurs starting from the gate. Therefore, with the technology disclosed in Patent Document 1, the accuracy of burr detection may decrease depending on the shape of the mold and the position of the sensor. Therefore, there is a need for a technology that can detect burr occurrence with high accuracy regardless of the shape of the mold.

[0006] The present disclosure has been made in view of the above-described background, and an object of one aspect is to provide a technique for detecting the occurrence of burrs with high accuracy regardless of the shape of a mold. [Means for solving the problem]

[0007] According to one embodiment, there is provided a manufacturing apparatus for a resin molded product, the manufacturing apparatus comprising: a fixed mold, a movable mold, a first sensor provided near a gate of a cavity formed by the fixed mold and the movable mold, and a determination unit that detects the occurrence of burrs based on the clearance between the fixed mold and the movable mold detected by the first sensor. [Effects of the Invention]

[0008] According to an embodiment, it is possible to detect the occurrence of burrs with high accuracy regardless of the shape of the mold.

[0009] The above and other objects, features, aspects and advantages of the present disclosure will become apparent from the following detailed description of the disclosure taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of the appearance of a manufacturing apparatus 100 according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an example of a main configuration related to burr detection of the manufacturing apparatus 100. [Figure 3] 10 is a diagram showing an example of peripheral devices and circuits connected to a mold 103. FIG. [Figure 4] FIG. 2 is a diagram illustrating an example of a detailed circuit configuration of a determination unit 321. [Figure 5] FIG. 10 is a diagram showing an example of the transition of pressure applied to a mold 103. [Figure 6] FIG. 2 is an enlarged view of the periphery of a cavity 205 of a mold 103. [Figure 7] FIG. 2 is a diagram illustrating a first example of a sensor 206. [Figure 8] FIG. 2 is a diagram illustrating a second example of a sensor 206. [Figure 9] FIG. 10 is a diagram showing an example of a change in the output value of a sensor 206. [Figure 10] FIG. 2 is a diagram showing a first example of the arrangement of a sensor 206. [Figure 11] FIG. 10 is a diagram illustrating a second example of the arrangement of the sensor 206. [Figure 12] FIG. 10 is a diagram illustrating a third example of the arrangement of the sensor 206. [Figure 13] FIG. 4 is a diagram showing an example of the configuration of a learning device 420 for generating a model for detecting burrs in the manufacturing apparatus 100. [Figure 14] 10 is a flowchart showing an example of an internal processing procedure of a learning device 420. [Figure 15] FIG. 10 is a diagram showing an example of the configuration of an inference device 430 for detecting molding defects using a trained model in the molding process of the manufacturing device 100. [Figure 16] 10 is a flowchart showing an example of the internal processing procedure of the inference device 430. [Figure 17] FIG. 4 is a diagram illustrating an example of a neural network model used by a learning device 420. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the technical idea according to the present disclosure will be described while referring to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[0012] <A. Configuration of Manufacturing Apparatus for Resin Molded Product> First, referring to FIGS. 1 to 3, the configuration of the manufacturing apparatus for a resin molded product according to the present embodiment will be described. The manufacturing apparatus for a resin molded product according to the present embodiment manufactures a resin molded product by injection molding. Further, the manufacturing apparatus for a resin molded product according to the present embodiment detects burrs based on the output value of a sensor 206 (see FIG. 2) provided near the gate of the mold. Further, the manufacturing apparatus for a resin molded product according to the present embodiment can present or automatically execute improvement of the manufacturing process (adjustment of resin injection speed, mold temperature, resin temperature, etc.).

[0013] FIG. 1 is a diagram showing an example of the appearance of a manufacturing apparatus 100 according to the present embodiment. Referring to FIG. 1, the configuration of the manufacturing apparatus 100 and the procedure of injection molding will be described. The manufacturing apparatus 100 mainly includes a mold 103 and an injection unit 110. The mold 103 includes a movable mold 104 and a fixed mold 105. The movable mold 104 is driven by a motor 112. The injection unit 110 includes a hopper 101, a cylinder 102, and a motor 111.

[0014] The injection unit 110 melts the resin and pours the melted resin into the mold 103. The hopper 101 is set with a resin material. The resin material is, for example, resin pellets or the like. In some aspects, the hopper 101 may be manually replenished with resin or may be provided with a mechanism for automatically replenishing resin from an external tank or the like.

[0015] Cylinder 102 is connected to hopper 101 and serves as a passageway for transporting the resin that has flowed in from hopper 101 to mold 103. More specifically, screw 346 (see FIG. 3) is provided inside cylinder 102, and as screw 346 rotates, the resin inside cylinder 102 is transported to mold 103. Cylinder 102 is provided with heater 345 (see FIG. 3) for heating the resin to prevent it from solidifying.

[0016] The mold 103 is a mold into which resin is poured and deforms the resin into the shape of the product. The movable mold 104 is a mold driven by a motor 112, and the mold 103 opens and closes as the movable mold 104 moves. The movable mold 104 is provided with an ejector pin 208 (see FIG. 2). When the mold 103 opens (when the movable mold 104 moves), the ejector pin 208 pushes out the resin molded product attached to the mold 103 (movable mold 104). The fixed mold 105 is connected to the cylinder 102. For example, the resin flows from the fixed mold 105 through a sprue, a runner, and a gate into a cavity. The cavity is a gap in the mold where the product is molded, and is provided between the movable mold 104 and the fixed mold 105.

[0017] Motor 111 drives screw 346. In one aspect, manufacturing apparatus 100 may include motor 111 and a motor driver (not shown) that controls motor 111. In another aspect, manufacturing apparatus 100 may include motor 111 but not a motor driver. In this case, manufacturing apparatus 100 may be connected to an external motor driver.

[0018] The motor 112 drives the movable mold 104 via a power transmission component. In one aspect, the manufacturing apparatus 100 may include the motor 112 and a motor driver (not shown) that controls the motor 112. In another aspect, the manufacturing apparatus 100 may include the motor 112 but not the motor driver. In this case, the manufacturing apparatus 100 may be connected to an external motor driver.

[0019] Next, the procedure for resin molding will be described. First, the injection unit 110 pours heated molten resin (hereinafter, also referred to as "molten resin") into the mold 103. More specifically, the injection unit 110 heats and melts the resin material (pellets, etc.) in the hopper 101. The molten resin flows from the hopper 101 into the cylinder 102. The molten resin that has flowed into the cylinder 102 is injected by the tip 347 (see FIG. 3) of the screw 346 inside the cylinder 102 through a nozzle 348 (see FIG. 3) into the fixed mold 105. The screw 346 is rotated by the motor 111.

[0020] The molten resin injected into the fixed mold 105 passes through the sprue 203 (see FIG. 2). The molten resin that has passed through the sprue 203 flows into a cavity 205 (see FIG. 2) formed by the fixed mold 105 and the movable mold 104. In this specification, the term "cavity 205" encompasses the gap in the shape of the molded product formed between the movable mold 104 and the fixed mold 105, and the concave mold (usually the fixed mold 105). The convex mold (usually the movable mold 104) is also sometimes called a core.

[0021] The cavity 205 is divided into a runner, a gate 204, and a molded product section, and the molten resin is filled in this order: runner, gate 204, and molded product section. The molded product section is a gap shaped like the product to be molded from the resin. The runner is a passageway for the molten resin from the sprue 203 to the molded product section. The gate 204 is provided between the runner and the molded product section and serves as the entrance for the resin to the molded product section. The gate 204 may also have functions such as adjusting the inflow pressure, inflow speed, and temperature of the resin and preventing backflow. The cavity 205 may also have a gap called a gas vent (air vent) (not shown) for venting gas generated during resin filling to the outside. In some aspects, the mold 103 may also have a sensor (not shown) for detecting gas clogging in the gas vent (air vent). In this case, the determination unit 321 may detect burrs based on the output value (gas pressure, etc.) of the sensor that detects the gas blockage and the output value of the sensor 206 (displacement of the mold 103, i.e., the clearance (gap) between the fixed mold 105 and the movable mold 104). For example, the determination unit 321 may determine that burrs have occurred when the output value of the sensor that detects the gas blockage exceeds (or falls below) a predetermined first threshold (gas blockage is detected) and the output value of the sensor 206 exceeds (or falls below) a predetermined second threshold (clearance is detected). As another example, the determination unit 321 may determine that burrs have occurred when the output value of the sensor that detects the gas blockage and the output value of the sensor 206 each exceed (or fall below) a first and second threshold (both gas blockage and clearance are detected).

[0022] After the cavity 205 is filled with molten resin, the molten resin comes into contact with the surfaces of the mold 103 (the movable mold 104 and the fixed mold 105), whereby the resin is cooled and solidified, and a molded product is formed in the cavity 205. After the resin has solidified, the movable mold 104 is driven by the motor 112 to separate from the fixed mold 105, and the mold 103 opens. At this time, the molded product generally adheres to the movable mold 104. An ejector pin 208 (see FIG. 2) provided on the movable mold 104 pushes the molded product in the direction (closing direction) opposite to the operating direction (opening direction) of the movable mold 104, thereby peeling the molded product from the movable mold 104. The peeling of the molded product from the movable mold 104 is also referred to as demolding.

[0023] The individual steps in injection molding can be divided into a step of closing the movable mold 104 and the fixed mold 105 at their parting surfaces 209 (see Figure 2) (mold clamping step), a step of pouring resin into the mold 103 (injection step), a step of pressurizing the resin at a constant pressure using the mold 103 (pressure holding step), a step of cooling the resin (cooling step), a step of opening the mold 103 (mold opening step), and a step of pushing the molded product out of the mold 103 (movable mold 104) using the ejector pins 208 (mold release step).

[0024] 2 is a diagram showing an example of a main configuration related to burr detection of the manufacturing apparatus 100. The manufacturing apparatus 100 includes a sensor 206 and a cable hole 210 on the mold 103 as main components related to burr detection.

[0025] The sensor 206 detects the displacement of the mold 103 indicating how much the mold 103 has deformed or moved due to the pressure inside the mold 103. The displacement of the mold 103 here indicates, for example, the clearance between the fixed-side mold 105 and the movable-side mold 104 caused by the movement of the movable-side mold 104 and / or the deformation of each mold (the movable-side mold 104 and / or the fixed-side mold 105). The sensor 206 may detect the displacement of the mold 103 (i.e., the clearance between the fixed-side mold 105 and the movable-side mold 104) or the pressure applied to the mold 103 (the movable-side mold 104 and / or the fixed-side mold 105). That is, the output value of the sensor 206 indicates the displacement of the mold 103 (the clearance between the fixed-side mold 105 and the movable-side mold 104) or the pressure applied to the mold 103. In a certain aspect, the sensor 206 may be an electromagnetic induction displacement sensor or a laser sensor. Details of each sensor will be described with reference to FIGS. 7 and 8.

[0026] The cable hole 210 is a hole for passing the cable of the sensor 206. The cable of the sensor 206 passes through the cable hole 210 and is connected to the amplifier 320 (see FIG. 3) or the determination unit 321 (see FIG. 3). The manufacturing apparatus 100 (the determination unit 321) detects the generation of burrs on the molded product based on the output value of the sensor 206 (the amount of displacement of the mold 103, i.e., the clearance between the fixed-side mold 105 and the movable-side mold 104) provided in the vicinity of the gate 204 (hereinafter, the vicinity of the gate 204 is referred to as the "gate vicinity").

[0027] <B. Circuit Configuration> Next, referring to FIGS. 3 and 4, the circuit related to burr detection in the manufacturing apparatus 100 will be described.

[0028] 3 is a diagram showing an example of peripheral devices and circuits connected to the mold 103. The manufacturing apparatus 100 includes, as main circuits related to burr detection, an amplifier 320, a determination unit 321, a display device 322, and a mold clamping detection sensor 331. The manufacturing apparatus 100 also includes the mold 103, an injection unit 110, and a control unit 340. The injection unit 110 includes a hopper 101, a cylinder 102, a heater 345, a screw 346, and an injection motor 342.

[0029] The injection unit 110 injects resin into the mold 103. The hopper 101 is a location where resin material such as pellets is introduced. The cylinder 102 melts the resin supplied from the hopper 101 using an internal screw 346 and transports the resin to the mold 103. The mold clamping detection sensor 331 outputs a mold clamping completion signal 323 when the mold 103 closes (when the movable mold 104 and the fixed mold 105 come into contact at the parting surface 209). The mold clamping completion signal 323 can be used by the determination unit 321 to determine the timing of execution of a flash detection process. The control unit 340 controls the overall operation of the manufacturing apparatus 100 (such as the operation of the mold 103 and the operation of the injection unit 110).

[0030] The amplifier 320 amplifies the output signal of the sensor 206 and the mold clamping completion signal 323, and outputs the amplified signals to the determination unit 321. As an example, the amplifier 320 may be realized by an operational amplifier element and its peripheral circuitry, etc.

[0031] The determination unit 321 acquires the output values ​​(signals) of each sensor received from the amplifier. The determination unit 321 detects the occurrence of burrs based on the output values ​​of the sensors 206. More specifically, the determination unit 321 detects the occurrence of burrs based on the displacement of the mold 103 (the clearance between the fixed mold 105 and the movable mold 104) indicated by the output values ​​of the sensors 206. Furthermore, the determination unit 321 may generate information on the cause of the burrs and information for improving the molding conditions (at least some or all of information on mold temperature adjustment, resin temperature adjustment, resin injection speed adjustment, etc.) based on the detection of the occurrence of burrs. The determination unit 321 outputs information including some or all of the information on the determination of the presence or absence of burrs, the identification of the cause of the burrs, and the improvement of the molding conditions to the display device 322. In some aspects, the determination unit 321 may generate a feedback signal (information on improvement of molding conditions) based on the burr determination result and transmit the feedback signal to the control unit 340. The control unit 340 may change or improve the molding conditions (mold temperature, resin temperature, resin injection speed, etc.) based on the feedback signal. As an example, the determination unit 321 may change or improve the molding conditions (mold temperature, resin temperature, resin injection speed, etc.) based on the difference between the output value of each of the one or more sensors 206 and a predetermined threshold value.

[0032] Furthermore, by acquiring the mold clamping detection signal, the determination unit 321 can easily detect the timing when the mold 103 is clamped. As a result, the manufacturing apparatus 100 does not need to perform burr detection processing in all states during injection molding, such as the injection molding measurement process, mold clamping process, injection process, pressure holding process, cooling process, molded product removal process, and mold opening process, and can perform burr detection processing only for a certain period of time (such as only the injection process and pressure holding process) after acquiring the mold clamping detection signal. In other words, the determination unit 321 can acquire the output value of the sensor 206 based on the timing when the mold clamping detection signal is acquired.

[0033] Display device 322 displays the information acquired from determination unit 321. In one aspect, determination unit 321 may be a personal computer (PC), and display device 322 may be a display. In another aspect, determination unit 321 may be a device equipped with a microcomputer, a system on chip (SoC), or the like. In this case, display device 322 may be any device, such as a display, tablet, or PC, that can receive and display information or video from determination unit 321.

[0034] In one aspect, some or all of amplifier 320, determination unit 321, and display device 322 may be realized as separate devices or components, or may be realized as an integrated device. In another aspect, some or all of amplifier 320, determination unit 321, and display device 322 may be incorporated into manufacturing apparatus 100.

[0035] 4 is a diagram showing an example of a detailed circuit configuration of the determination unit 321. The determination unit 321 includes a processor 401, a memory 402, a storage 403, an external device IF (Interface) 404, an input IF 405, an output IF 406, and a communication IF 407. These circuits are connected to each other by a bus 408.

[0036] The processor 401 may execute a program for implementing various functions of the determination unit 321. The processor 401 may be configured, for example, by at least one integrated circuit. The integrated circuit may be configured, for example, by at least one central processing unit (CPU), at least one graphics processing unit (GPU), at least one field programmable gate array (FPGA), at least one application specific integrated circuit (ASIC), or a combination thereof.

[0037] Memory 402 stores programs executed by processor 401 and data referenced by processor 401. In one aspect, memory 402 may be realized by a dynamic random access memory (DRAM), a static random access memory (SRAM), or the like.

[0038] Storage 403 is a non-volatile memory that stores programs executed by processor 401 and data referenced by processor 401. In this case, processor 401 executes programs read from storage 403 to memory 402 and references data read from storage 403 to memory 402. In one aspect, storage 403 may be realized by a hard disk drive (HDD), a solid state drive (SSD), an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a flash memory, or the like.

[0039] The external device IF 404 can be connected to any external device. The external device IF 404 is connected to the amplifier 320 and receives the output values ​​of each sensor via the amplifier 320. The external device IF 404 writes the received output values ​​of each sensor into the memory 402 or the storage 403. In one aspect, the external device IF 404 may be realized by a USB (Universal Serial Bus) terminal or the like.

[0040] The input IF 405 may be connected to any input device, such as a keyboard, mouse, touchpad, or gamepad. In one aspect, a user may input an on / off command for the burr detection monitoring function to the determination unit 321 via the input device 410 connected to the input IF 405. In another aspect, the user may select one or more improvements to the molding conditions displayed on the determination unit 321 via the input device 410, or input an instruction to execute the improvements to the determination unit 321. In this case, the determination unit 321 may output parameters (feedback signals) of the selected improvements to the control unit 340 of the manufacturing apparatus 100. The control unit 340 may adjust the molding conditions based on the parameters (feedback signals). In one aspect, the input IF 405 may be implemented by a USB terminal, a PS / 2 terminal, a Bluetooth (registered trademark) module, or the like.

[0041] Output IF406 may be connected to any output device, such as a cathode ray tube display, a liquid crystal display, or an organic electroluminescence (EL) display. Output IF406 is connected to display device 322. Output IF406 outputs information to display device 322, including the presence or absence of burrs, the cause of the burrs, and some or all of the improvement measures for molding conditions. In one aspect, output IF406 may be implemented by a USB terminal, a D-sub terminal, a DVI (Digital Visual Interface) terminal, an HDMI (registered trademark) (High-Definition Multimedia Interface) terminal, or the like. In one aspect, when display device 322 is a PC or the like, determination unit 321 may be connected to display device 322 via external device IF404 or communication IF407.

[0042] The communication IF407 is connected to a wired or wireless network device. The communication IF407 can be connected to an external learning device 420 and an inference device 430. In a certain aspect, the communication IF407 may be realized by a wired LAN (Local Area Network) port, a Wi-Fi (registered trademark) (Wireless Fidelity) module, etc. In other aspects, the communication IF407 may transmit and receive data using communication protocols such as TCP / IP (Transmission Control Protocol / Internet Protocol) and UDP (User Datagram Protocol).

[0043] The learning device 420 takes in the output values of the pre-prepared sensor 206 and the determination result of the burr as learning data, and generates a learned model for burr determination by machine learning. In a certain aspect, the determination unit 321 may transmit the burr determination result to the learning device 420 as learning data via the communication IF407.

[0044] The inference device 430 uses the learned model to output the presence or absence of burrs, the cause of burrs, and improvement measures for molding conditions based on the output value of the sensor 206 obtained from the determination unit 321. In this case, the determination unit 321 may receive the presence or absence of burrs, the cause of burrs, and improvement measures for molding conditions from the inference device 430 via the communication IF407, and display this information on the display device 322. In a certain aspect, instead of analyzing the signal obtained from the amplifier 320 by itself, the determination unit 321 may transfer it to the inference device 430 and receive the presence or absence of burrs, the cause of burrs, and improvement measures for molding conditions from the inference device 430. In other aspects, the determination unit 321 may output the presence or absence of burrs, the cause of burrs, and improvement measures for molding conditions using the learned model received from the learning device 420.

[0045] <C. Timing of Burr Generation and Mechanism of Burr Detection> Next, referring to FIGS. 5 and FIGS. 6, the timing of burr generation and the mechanism of burr detection will be described.

[0046] (a. Timing of burr occurrence) 5 is a diagram showing an example of the transition of the pressure applied to the mold 103. The pressure applied to the mold 103 changes with each step of the injection molding. Graph 500 shows the transition of the pressure applied to the mold 103 through a series of steps of the injection molding.

[0047] Section 501 shows the pressure inside the mold 103 during the injection process in the series of steps for manufacturing a resin molded product. During the injection process, as the resin fills the cavity 205, the internal pressure of the mold 103 rises and reaches a peak. Once the resin has completely filled the cavity 205, the manufacturing process for a resin molded product transitions to a pressure holding process.

[0048] Section 502 shows the pressure inside the mold 103 during the pressure holding process in the series of manufacturing processes for a resin molded product. During the pressure holding process, the internal pressure of the mold 103 becomes constant. Once the application of pressure to the resin (pressure holding) is completed, the manufacturing process for a resin molded product moves to the cooling process.

[0049] Section 503 shows the pressure inside the mold 103 during the cooling process. As the resin cools, the pressure inside the mold 103 decreases. Once the resin has cooled, the manufacturing process for the resin molded product moves to the mold opening process and demolding process.

[0050] As shown in graph 500, the pressure applied to mold 103 is greatest during the injection process (section 501) and the pressure holding process (section 502). Therefore, it is desirable for determination unit 321 to detect the occurrence of burrs based on the output value of sensor 206 during the injection process (section 501) and the pressure holding process (section 502).

[0051] (b. Causes of burrs) Next, the causes of burrs occurring in the injection process (section 501) and the pressure holding process (section 502) will be described.

[0052] In the injection process (section 501), molten resin is forcefully injected from the cylinder 102 into the cavity 205. As a result, the pressure inside the mold 103 increases due to the resin flowing into the cavity 205. When the pressure inside the mold 103 exceeds the clamping force of the mold 103, the movable mold 104 moves back, creating a clearance at the parting surface 209. When resin flows into this clearance, flash occurs on the molded product.

[0053] In the pressure holding step (section 502), a constant pressure is applied to the mold 103 by the motor 112 until the resin in the cavity 205 hardens. At this time, if the internal pressure of the mold 103 (pressure received from the resin in the cavity 205) exceeds the clamping force of the mold 103, the movable mold 104 moves back, and a clearance is generated at the parting surface 209. Resin flows into this clearance, causing flash to form on the molded product.

[0054] Additionally, during the injection process (section 501) and the pressure holding process (section 502), the mold 103 (movable mold 104 and / or fixed mold 105) may be deformed due to the pressure inside the mold 103. When the mold 103 is deformed, a gap may occur at the parting surface 209, and resin may flow out from the gap, causing flash on the molded product.

[0055] Furthermore, burrs can occur suddenly due to a wide range of factors, such as variations between shots (variations in resin temperature, resin injection speed, etc.), variations in material properties between resin lots (variations in viscosity, etc.), and wear or deterioration of the mold.

[0056] Therefore, by disposing the sensor 206 near the gate where the clearance inside the mold 103 is most likely to occur, the manufacturing apparatus 100 can detect the occurrence of the clearance with high accuracy. As a result, the manufacturing apparatus 100 can reduce the number of steps required for frequent burr inspection and deburring work.

[0057] (c. Burr detection mechanism) Fig. 6 is an enlarged view of the periphery of cavity 205 of mold 103. In the example of Fig. 6, movable mold 104 and fixed mold 105 are mated to form parting surface 209.

[0058] Resin injected from the injection unit 110 into the mold 103 passes through the sprue 203 and flows into the cavity 205. More specifically, the resin passes through the runner and gate 204 in the cavity 205 and flows into the molded product portion. When the resin fills the cavity 205 until its volume is approximately the same as the volume of the cavity 205, the pressure inside the cavity 205 (inside the mold 103) increases. When this pressure exceeds the clamping force of the mold 103, a clearance occurs at the parting surface 209. Clearance is particularly likely to occur around the gate.

[0059] The sensor 206 is disposed near the gate where clearance is likely to occur. In one aspect, the sensor 206 may be embedded in a position near the gate of the mold 103. In another aspect, the sensor 206 may be embedded in the movable mold 104 or the fixed mold 105. The cable hole 210 is provided in the mold in which the sensor 206 is embedded. In another aspect, the sensor 206 may be provided on a side surface of the mold 103 (such as the top surface, bottom surface, or side surface of the mold 103). In this case, it is also desirable to dispose the sensor 206 as close to the gate as possible.

[0060] Gate 204 is the one where the pressure is most likely to rise and clearance is most likely to occur. That is, Gate 204 tends to have a larger clearance than other parts of the mold 103. Therefore, when clearance occurs at the parting surface 209, the amount of clearance near the gate tends to be larger than the amount of clearance at a position far from Gate 204. By arranging the sensor 206 near the gate, the output value of the sensor 206 will be greatly displaced when clearance occurs. The manufacturing apparatus 100 can detect the occurrence of clearance with high precision (or high sensitivity) by checking the displacement of the output value of the sensor 206 arranged near the gate, even if only a slight clearance has occurred.

[0061] <D. Variations in Sensor Type or Arrangement> Next, referring to FIGS. 7 to 12, variations in the type and arrangement of the sensor 206 in the manufacturing apparatus 100 according to the present embodiment will be described. Note that the configurations shown in FIGS. 7 to 12 may be used in appropriate combinations. Also, each configuration shown in FIGS. 7 to 12 and combinations thereof can be arbitrarily combined and used with the configurations and technologies described with reference to FIGS. 1 to 6.

[0062] (a. Sensor Type) FIG. 7 is a diagram showing a first example of the sensor 206. In the example of FIG. 7, the sensor 206 is an electromagnetic induction type displacement sensor. The electromagnetic induction type displacement sensor is a sensor embedded in the mold 103 and is composed of a conversion circuit, a primary winding coil, a secondary winding coil, and a magnetic body that moves inside the coil. The electromagnetic induction type displacement sensor generates an alternating magnetic field by exciting the primary winding coil with a reference primary alternating current signal. Then, the electromagnetic induction type displacement sensor detects an induction output corresponding to the positions of two sets of coils and three magnetic bodies in the alternating magnetic field, and measures the position of the magnetic body from the phase difference from the initial state.

[0063] The electromagnetic induction type displacement sensor can be installed near the cavity (particularly the gate 204) of the mold 103, and can detect the clearance generated at the parting surface 209 near the molded product or the gate 204. Therefore, by providing the electromagnetic induction type displacement sensor near the gate, the manufacturing apparatus 100 can detect burrs with high accuracy.

[0064] The sensor 206 (electromagnetic induction type displacement sensor) has a button with a convex shape, etc. The button of the sensor 206 sinks normally when the movable mold 104 and the fixed mold 105 are in close contact at the parting surface 209. When a clearance occurs at the parting surface 209, the amount of depression of the button of the sensor 206 changes.

[0065] The sensor 206 is disposed near the gate where pressure is most likely to increase and clearance is most likely to occur within the mold 103. Therefore, even if only a small clearance occurs, the amount of depression of the button varies significantly. The sensor 206 varies the voltage and / or current it outputs in accordance with the variation in the amount of depression of the button. In some aspects, the sensor 206 (electromagnetic induction type displacement sensor) may be embedded in the movable mold 104 or the fixed mold 105. In either case, it is preferable that the sensor 206 be located near the gate. It is also desirable that the tolerance range of the sensor 206 be as small as possible, or that the influence of the tolerance be minimal. By embedding the sensor 206 near the gate (where the clearance is large), the influence of the tolerance of the sensor 206 can be reduced.

[0066] FIG. 8 is a diagram showing a second example of the sensor 206. In the example of FIG. 8, the sensor 206 is a laser sensor. A laser sensor emits light using a light-emitting element in a light-emitter (outputs a laser) and receives light using a light-receiving element in a light-receiver (detects the laser). The laser sensor detects displacement (mold clearance) by utilizing the time difference between this emission and reception. Examples of laser sensors include a type in which the light-emitter and light-receiver are integrated, a type in which the light-emitter and light-receiver are separate, and a transmission type in which the light-emitter and light-receiver face each other with a detection object between them. The manufacturing apparatus 100 may be equipped with any type of laser sensor. The laser sensor does not need to be embedded inside the mold, but may be installed on the side (top, bottom, side, etc.) of the mold 103.

[0067] The sensor 206 includes a light projector and a light receiver. As an example, the light projector may be provided in the movable mold 104, and the light receiver may be provided in the fixed mold 105. As another example, the light receiver may be provided in the fixed mold 105, and the light projector may be provided in the movable mold 104. When the sensor 206 is an integrated laser sensor, the light projector and the light receiver may be provided in the movable mold 104, or the light projector and the light receiver may be provided in the fixed mold 105.

[0068] The sensor 206 is disposed near the gate where pressure is most likely to increase and clearance is most likely to occur within the mold 103. Therefore, even if only a small clearance occurs, the distance 810 between the light emitter and the light receiver varies significantly, and the time difference between when the laser 801 emits light and when it is received also varies significantly. The sensor 206 varies the voltage and / or current it outputs in accordance with the variation in the time difference between when the laser 801 emits light and when it is received. In one aspect, the sensor 206 (laser sensor) may be embedded in the mold 103 or may be installed on a side surface (top surface, bottom surface, side surface, etc.) of the mold 103. In either case, the sensor 206 is preferably located near the gate.

[0069] FIG. 9 is a diagram showing an example of the change in the output value of the sensor 206. Graph 900A shows the change in the output value of the sensor 206 when a molded product without flash is molded. Graph 900B shows the change in the output value of the sensor 206 when a molded product with flash is molded. The horizontal axis represents time (s (seconds)), and the vertical axis represents clearance (μm (micrometers)). Note that the units of each axis are merely examples, and the output value of the sensor 206 may represent clearance in any unit at any time interval. In some aspects, the determination unit 321 may calculate the clearance based on the output value (voltage, current, or digital value) of the sensor 206. Whether an electromagnetic induction displacement sensor or a laser sensor is used as the sensor 206, the change in the output value of the sensor 206 has a shape as shown in FIG. 9.

[0070] According to graph 900A, even if no burrs are generated, the clearance is L at point 910A. A and is not 0. This is because the movable mold 104 and the fixed mold 105 are curved (deformed) due to the pressure inside the mold 103. The amount of deformation is maximum at the gate 204. Clearance L A is a clearance that does not cause burrs.

[0071] According to graph 900B, at point 910B, clearance L A Larger clearance than L B This is because the mold 103 is bent more than usual due to the internal pressure of the mold 103 being higher than expected. When the internal pressure of the mold 103 increases in this way, the clearance near the gate also changes the most.

[0072] As an example, the determination unit 321 may store a threshold value of the clearance for determining whether or not there is a burr in the storage 403. The determination unit 321 may determine whether or not the clearance value acquired from the sensor 206 (or the clearance value calculated from the output value of the sensor 206) exceeds the threshold value. In the example of FIG. 9, the determination unit 321 determines whether or not the clearance L A is below the threshold value, it is determined that no burrs have occurred, and the clearance L B exceeds the threshold value, it can be determined that a burr has occurred.

[0073] (b. Sensor placement) 10 to 12, variations in the arrangement of the sensors 206 will be described. Depending on the shape of the mold 103, the arrangement locations and / or the number of sensors 206 may be changed.

[0074] FIG. 10 is a diagram showing a first example of the arrangement of the sensor 206. One or more sensors 206 may be arranged in the movable mold 104 and / or the fixed mold 105. The position of the sensor 206 may be changed as appropriate depending on the shape of the mold 103. An example of the arrangement of the sensor 206 will be described with reference to FIG. 10. Molds 1010, 1020, and 1030 are variations of the movable mold 104, and show the movable mold 104 as seen from the parting surface 209. Note that, although the gate 204 and the cavity 205 are separated in the figure, the gate 204 may be included in the cavity 205. A runner may be present between the sprue 203 and the gate 204.

[0075] The mold 1010 has one sensor 206 near the gate 204. In the example of the mold 1010, the sensor 206 is provided on the central axis of the mold 1010 when viewed from the left and right, but the location of the sensor 206 is not limited to this. In some aspects, the sensor 206 may be positioned closer to the gate 204, shifted to the left or right.

[0076] The mold 1020 includes a plurality of sensors 206 near the gate 204. In the example of the mold 1020, two sensors 206 are arranged on the left and right sides as viewed from the gate 204, but the number and locations of the sensors 206 are not limited to this. In one aspect, the sensors 206 may be arranged above and below the gate 204, or any number of sensors 206 may be arranged in a ring, or any number of sensors 206 may be arranged in any location near the gate 204.

[0077] In one aspect, when there are multiple sensors 206 near the gate, the determination unit 321 may determine that a burr has occurred based on the output value exceeding a threshold value of any one of the multiple sensors 206. In another aspect, when there are multiple sensors 206 near the gate, the determination unit 321 may determine that a burr has occurred based on the output values ​​of at least a predetermined number of sensors 206 or all of the multiple sensors 206 exceeding a threshold value.

[0078] In the mold 1030, a plurality of sensors 206 are arranged near the gate and at positions far from the gate 204. As an example, one or more sensors 206 may be provided near the gate, and one or more sensors 206 may be provided near the end of the molded product portion (near a thin portion of the resin far from the gate 204 of the molded product portion).

[0079] In one aspect, when one sensor 206 is disposed near the gate and one sensor 206 is disposed far from the gate 204, the determination unit 321 may determine whether or not a burr is present based on the difference in output values ​​of the sensors 206. In another aspect, when multiple sensors 206 are disposed near the gate and one sensor 206 is disposed far from the gate 204, the determination unit 321 may determine whether or not a burr is present based on the difference between the average value of the output values ​​of the sensors 206 near the gate and the average value of the output values ​​of the sensors 206 far from the gate 204.

[0080] FIG. 11 is a diagram showing a second example of the arrangement of the sensor 206. A mold 1100 is an example of the movable mold 104, and shows the movable mold 104 as seen from the parting surface 209. Note that the interior of the cavity 205 may be divided into a runner, a gate 204, and a molded product portion. In the example shown in FIG. 11, the location where the sensor 206 can be arranged is determined based on the distance from the gate 204 of the movable mold 104 and / or the fixed mold 105. As an example, the location where the sensor 206 can be arranged is an area 1110 within a predetermined distance from the gate 204. In one aspect, the area 1110 may be within a circle having a predetermined radius from the center of the gate 204. For example, the area 1110 may be determined based on the diameter (gate diameter) of the gate 204. The area 1110 may be a circle having a diameter (diameter or radius) whose ratio to the gate diameter (diameter or radius) is a predetermined ratio. In another aspect, the region 1110 may be rectangular or any other shape. The region 1110 (the limit distance that the sensor 206 may be away from the gate 204) may be determined by a previous experiment on the sensitivity of the sensor 206, or the like. The region 1110 (the limit distance that the sensor 206 may be away from the gate 204) is a region on the parting surface 209 where a clearance is likely to occur between the movable mold 104 and the fixed mold 105, and where the sensor 206 is likely to detect the occurrence of burrs. By arranging one or more sensors 206 within a predetermined distance from the gate 204, the manufacturing apparatus 100 can detect the occurrence of burrs with high accuracy (high sensitivity).

[0081] 12 is a diagram showing a third example of the arrangement of the sensor 206. A mold 1200 is an example of the movable mold 104, and shows the movable mold 104 as seen from the parting surface 209. In the example shown in FIG. 12, the sensor 206 is arranged near the gate and near the tie bar 1210. The tie bar 1210 is a support that guides the opening and closing of the mold 103 provided in the manufacturing apparatus 100, and serves to receive the internal pressure of the mold 103.

[0082] As described above, since clearance is likely to occur near the gate, the manufacturing apparatus 100 can easily detect burrs near the gate with high precision (high sensitivity). However, the farther away from the gate 204, the less likely clearance is to occur. Therefore, although the sensor 206 can detect the generation of burrs with high precision near the gate, it becomes difficult to detect the generation of burrs at locations far from the gate 204. Thus, in order to also detect burrs at positions far from the gate 204 with high precision, the manufacturing apparatus 100 may be provided with the sensor 206 not only near the gate but also near the tie bar 1210. By doing so, the manufacturing apparatus 100 can confirm the displacement of the output values of the plurality of sensors 206 at positions far from the gate 204 (near the tie bar 1210) (the displacement of the mold 103, that is, the clearance between the fixed-side mold 105 and the movable-side mold 104), and can thus detect the presence or absence of burrs outside the gate vicinity with high precision (high sensitivity).

[0083] In a certain aspect, the arrangement methods of the sensors 206 shown in FIGS. 10 to 12 may be used in combination. Also, in other aspects, depending on the type of the sensor 206 or the shape of the mold, the sensor 206 may be arranged on both or either one of the movable-side mold 104 and the fixed-side mold 105.

[0084] <E. Application Using Machine Learning> Next, referring to FIGS. 13 to 17, an example of incorporating machine learning into the manufacturing apparatus 100 according to the present embodiment to perform burr detection and improvement of molding conditions will be described.

[0085] 13 is a diagram showing an example of the configuration of a learning device 420 for generating a model for detecting burrs in the manufacturing apparatus 100. The learning device 420 mainly includes a data acquisition unit 1310, a model generation unit 1320, and a trained model storage unit 1330. In some aspects, some or all of the data acquisition unit 1310, the model generation unit 1320, and the trained model storage unit 1330 may be realized as a program. In this case, for example, the learning device 420 may include a processor (not shown), a memory (not shown), and a storage (not shown), and may execute the program on this hardware.

[0086] The data acquisition unit 1310 acquires, as learning data, information associating the output value of the sensor 206 with information (answer) on the presence or absence of burrs. The information may also include the molding conditions of the resin. The output value of the sensor 206 indicates, for example, the output value of one or more sensors 206 provided in the movable mold 104 and / or the fixed mold 105. As an example, the manufacturing apparatus 100 may have a mold 103 provided with one or more sensors 206 as shown in FIGS. 10 to 12.

[0087] The model generation unit 1320 uses the learning data output from the data acquisition unit 1310 to learn how to determine the presence or absence of burrs based on the output values ​​of each sensor 206, estimate (identify) the cause of burrs, and adjust molding conditions when burrs occur. The model generation unit 1320 also learns how to adjust molding conditions (mold temperature, resin temperature, resin injection speed, etc.) based on combinations of the output values ​​of the sensors 206, molding conditions, and the presence or absence (answer) of burrs.

[0088] In one aspect, learning data may be input by a user via a medium or a network to learning device 420. In another aspect, learning device 420 may receive, as learning data, data used in determining the presence or absence of a past molding defect from determination unit 321 via the network.

[0089] In another aspect, learning device 420 may be realized in any form, such as a server, a virtual machine in a cloud environment, a container, an application, etc. In another aspect, learning device 420 may be realized as an integrated unit with a manufacturing device or a determination unit.

[0090] The model generation unit 1320 may use any algorithm such as supervised learning, unsupervised learning, reinforcement learning, etc. Upon completion of learning, the model generation unit 1320 outputs a trained model and stores the trained model in the trained model storage unit 1330.

[0091] 14 is a flowchart showing an example of the procedure of internal processing of learning device 420. In one aspect, the processor of learning device 420 may read a program for executing the processing shown in FIG. 14 from storage onto a memory, and execute the program.

[0092] In step S1410, the learning device 420 acquires learning data. The learning data is, for example, information that associates molding conditions, output values ​​of one or more sensors 206, and information (answer) on the presence or absence of burrs. The learning device 420 may input the learning data from a user via a medium or a network, or may receive data used in determining the presence or absence of past molding defects from the manufacturing apparatus 100 as learning data.

[0093] In step S1420, the learning device 420 executes a learning process using the learning data by the model generation unit 1320. When the learning is completed, the model generation unit 1320 outputs the learned model.

[0094] In step S1430, the learning device 420 stores the trained model in the trained model storage unit 1330. In one aspect, when the learning device 420 generates a new trained model, the learning device 420 may transmit the trained model to the inference device 430. In another aspect, the learning device 420 may transmit the trained model to the inference device 430 based on receiving a request from the inference device 430.

[0095] 15 is a diagram showing an example of the configuration of an inference device 430 for detecting molding defects using a trained model in the molding process of the manufacturing apparatus 100. The inference device 430 mainly includes a data acquisition unit 1510, an inference unit 1520, and a trained model storage unit 1530. In some aspects, some or all of the data acquisition unit 1510, the inference unit 1520, and the trained model storage unit 1530 may be realized as programs. In this case, for example, the inference device 430 may include a processor (not shown), a memory (not shown), and a storage (not shown), and may execute the program on this hardware.

[0096] The data acquisition unit 1510 acquires, as input for inference, output values ​​of one or more sensors 206 provided on the mold 103. In other words, the data acquisition unit 1510 acquires data similar to the data acquired by the determination unit 321. The data acquisition unit 1510 may also acquire information on molding conditions from the manufacturing apparatus 100 in order to adjust the molding conditions.

[0097] The inference unit 1520 inputs the data transferred from the data acquisition unit 1510 to the trained model generated by the model generation unit 1320. The trained model outputs an inference result based on the input data. The inference result is a determination result of whether or not a burr is present, an estimation result of the cause of the burr, adjustment information for molding conditions, etc. The inference unit 1520 outputs the inference result to the determination unit 321. In one aspect, the inference unit 1520 may output only the determination result of whether or not a burr is present. In another aspect, the inference unit 1520 may output an estimation result of the cause of the burr and adjustment information for molding conditions together with the determination result of whether or not a burr is present.

[0098] 9 and other figures, the determination unit 321 compares the output value of the sensor 206 with a threshold value, but the inference unit 1520 uses the output value of the sensor 206 as an input to a trained model. Therefore, if the determination unit 321 has an inference device 430 built in or requests an external inference device 430 to perform the determination process, the determination unit 321 does not need to store a threshold value.

[0099] The trained model storage unit 1530 stores the trained model. In one aspect, the inference device 430 may periodically or irregularly receive the trained model from the learning device 420 and store the trained model in the trained model storage unit 1530. In another aspect, the trained model storage unit 1530 may be the same storage unit as the trained model storage unit 1330.

[0100] In one aspect, learning device 420 and inference device 430 may be the same device. In another aspect, inference device 430 may be implemented in any form, such as a server, a virtual machine in a cloud environment, a container, or an application. In another aspect, inference device 430 may be implemented as an integrated unit with a manufacturing device or a determination unit.

[0101] 16 is a flowchart showing an example of an internal processing procedure of inference device 430. In one aspect, the processor of inference device 430 may read a program for executing the processing shown in FIG. 16 from storage onto memory and execute the program.

[0102] In step S1610, the inference device 430 acquires information used to detect burrs and identify the cause of the burrs. This information is the same as the information used by the determination unit 321 to detect burrs and identify the cause of the burrs. In one aspect, the inference device 430 may acquire information (output values ​​of one or more sensors 206 provided in the manufacturing apparatus) directly or via the amplifier 320. In another aspect, the inference device 430 may acquire information (output values ​​of one or more sensors 206 provided in the manufacturing apparatus) from the determination unit 321. In another aspect, the inference device 430 may acquire molding conditions from the control unit 340 or the determination unit 321. The inference device 430 may use the acquired molding conditions to generate parameters for improving the molding conditions.

[0103] In step S1620, inference device 430 inputs information used to detect burrs and identify the cause of burrs into the trained model.

[0104] In step S1630, inference device 430 outputs data (inference results). The data (inference results) may include the results of burr detection, the results of identifying the cause of burr, parameters for improving molding conditions, etc.

[0105] In step S1640, inference device 430 detects the occurrence of burrs by referring to the inference result. In one aspect, inference device 430 may transmit the inference result to determination unit 321. In another aspect, inference device 430 may transmit a burr occurrence notification or the like to determination unit 321 by referring to the inference result. Note that determination unit 321 may have the functions of inference device 430. In this case, determination unit 321 executes the processing procedure shown in FIG. 16 by using a trained model instead of making a determination using a threshold value.

[0106] FIG. 17 is a diagram showing an example of a neural network model used by the learning device 420. The neural network model is used for supervised learning. Supervised learning is a technique in which a learning device is provided with a set of input and result (label) data, and the device learns the features of the learning data and infers the result from the input. Note that the learning device 420 may perform learning using any technique other than a neural network.

[0107] A neural network consists of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer, or two or more layers.

[0108] For example, in a three-layer neural network as shown in FIG. 17, when multiple inputs are input as training data to the input layer (X1-X3), the values ​​of the training data are multiplied by weight W1 (w11-w16) and input to the intermediate layer (Y1-Y2), and the result is further multiplied by weight W2 (w21-w26) and output from the output layer (Z1-Z3). This output result varies depending on the values ​​of weights W1 and W2. The learning device 420 according to this embodiment inputs molding conditions, output values ​​of one or more sensors 206, etc., into the neural network as training data and repeatedly executes the learning process.

[0109] As described above, the manufacturing apparatus 100 according to this embodiment can detect the occurrence of burrs in a resin molded product by analyzing the output values ​​of one or more sensors 206 provided near the gate. Furthermore, the manufacturing apparatus 100 according to this embodiment can identify the cause of burrs by analyzing the output values ​​of one or more sensors 206 provided near the gate, and generate and output instructions for improving molding conditions and parameters for doing so. The technology of the present disclosure can also be applied to, for example, metal casting using a die-casting mold.

[0110] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present disclosure is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope equivalent to the claims. Furthermore, the disclosures described in the embodiments and each modification are intended to be implemented, as far as possible, either alone or in combination. [Explanation of symbols]

[0111] 100 Manufacturing equipment, 101 Hopper, 102 Cylinder, 103, 1010, 1020, 1030, 1100, 1200 Mold, 104 Movable side mold, 105 Fixed side mold, 110 Injection unit, 111, 112 Motor, 203 Sprue, 204 Gate, 205 Cavity, 206 Sensor, 208 Ejector pin, 209 Parting surface, 210 Cable hole, 320 Amplifier, 321 Judgment unit, 322 Display device, 331 Mold clamping detection sensor, 340 Control unit, 342 Injection motor, 345 Heater, 346 Screw, 347 Tip, 348 Nozzle, 401 Processor, 402 Memory, 403 Storage, 404 External device IF, 405 Input IF, 406 Output IF, 407 Communication IF, 408 bus, 410 input device, 420 learning device, 430 inference device, 500, 900A, 900B graph, 501, 502, 503 section, 801 laser, 810 distance, 910A, 910B point, 1110 area, 1210 tie bar, 1310, 1510 data acquisition unit, 1320 model generation unit, 1330, 1530 trained model memory unit, 1520 inference unit.

Claims

1. A fixed side mold; A movable mold; a first sensor provided near a gate of a cavity formed by the fixed mold and the movable mold; a determining unit that detects the occurrence of burrs based on the clearance between the fixed mold and the movable mold detected by the first sensor.

2. The apparatus for manufacturing a resin molded product according to claim 1 , wherein the first sensor is disposed within an area determined based on a gate diameter of the gate.

3. 3. The apparatus for manufacturing a resin molded product according to claim 2, wherein the region is a circle having a diameter whose ratio to the gate diameter is a predetermined ratio.

4. 4. The apparatus for manufacturing a resin molded product according to claim 1, wherein the first sensor is provided in the movable mold.

5. Further provided are a plurality of tie bars which serve as guides in the direction of movement of the movable mold, There are a plurality of the first sensors, a plurality of the first sensors are provided near each of the plurality of tie bars; 4. The manufacturing device for resin molded products according to claim 1, wherein detecting the occurrence of burrs based on the clearance between the fixed mold and the movable mold detected by the first sensor includes detecting the clearance between the fixed mold and the movable mold based on the output value of each of the first sensors provided near the gate and near each of the plurality of tie bars.

6. a data acquisition unit that acquires, as learning data, a clearance between a fixed mold and a movable mold obtained by a sensor provided near a gate of a mold of a manufacturing device for a resin molded product, molding conditions for the resin molded product, and a determination result of the presence or absence of burrs on the resin molded product; and a model generation unit that uses the learning data to generate a trained model for inferring the molding conditions and the presence or absence of burrs of the resin molded product.

7. a data acquisition unit that acquires, as input data, the clearance between the fixed mold and the movable mold obtained by the sensor provided near the gate of the mold from the manufacturing apparatus; An inference device comprising: an inference unit that inputs the input data into the trained model according to claim 6 and outputs information on the presence or absence of burrs on the resin molded product.

Images