Anomaly detection system, anomaly detection device, anomaly detection method, program, and trained model
The anomaly detection system in injection molding devices uses pressure sensor data to integrate and ratio pressure values, effectively detecting temporary gate clogging and ensuring consistent product quality by addressing the limitations of conventional methods.
Patent Information
- Application Number
- JP2020201072
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-12-03
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2040-12-03
AI Technical Summary
Existing anomaly detection systems in injection molding devices fail to reliably detect temporary gate clogging, which can lead to abnormalities in molded products due to variations in pressure during the molding process.
Anomaly detection system that utilizes pressure sensors to integrate pressure values over specific time intervals, calculating a ratio of these values to detect temporary gate clogging by analyzing the difference in pressure influence during the filling and holding phases.
Enhances the detection of temporary gate clogging, ensuring consistent product quality by identifying abnormalities that conventional methods miss, thereby improving the reliability of the molding process.
Smart Images

Figure 0007585748000001 
Figure 0007585748000002 
Figure 0007585748000003
Abstract
Description
[Technical field]
[0001] The present invention relates to an anomaly detection system, an anomaly detection device, an anomaly detection method, a program, and a trained model. [Background technology]
[0002] There is known an injection molding apparatus that supplies a molten liquid of a molding material to a cavity formed between a plurality of molds to form a molded product. For example, Patent Document 1 discloses a technique that detects the injection speed by an encoder attached to an injection servo motor, and stops the injection operation if the injection speed does not reach the monitored speed during a monitoring time of the injection process. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP2009-000929A Summary of the Invention [Problem to be solved by the invention]
[0004] In a molding device, if a gate becomes clogged, the molding material will not flow into the cavity as designed, which may cause defects in the molded product. The technology in Patent Document 1 detects gate clogging by utilizing the fact that, when the gate becomes clogged, the load pressure during injection becomes higher than the load pressure when molding a good product, and the injection speed becomes slower than the injection speed when molding a good product.
[0005] However, the inventors discovered that even if a gate is clogged, the load pressure during injection may not change much from when a good product was molded. In such cases, the technology of Patent Document 1 cannot detect the gate clogging.
[0006] Therefore, an object of the present invention is to provide an anomaly detection system, an anomaly detection device, an anomaly detection method, a program, and a trained model that can more reliably detect gate blockages. [Means for solving the problem]
[0007] (1) The abnormality detection system disclosed herein is an abnormality detection system including a molding apparatus for molding a molded product, and an abnormality detection device for detecting an abnormality in the molding apparatus, the molding apparatus including a mold section having therein a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates, an injection section which performs a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, and an abnormality detection device which detects an abnormality in the molding apparatus. or a pressure sensor detecting the pressure of the molding material at the gate, wherein the abnormality detection device has a data acquisition unit that acquires a first evaluation value obtained by integrating over time the pressure detected by the pressure sensor from a first point in time after the filling operation to a second point in time before the pressure-holding release operation and a second evaluation value obtained by integrating over time the pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the pressure-holding release operation, and an abnormality detection unit that acquires detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value.
[0008] As a result of intensive research, the inventors discovered that when temporary clogging occurs in some of the multiple gates, the pressure is detected by a pressure sensor installed between the spool and the gate, and the effect of the pressure rise is more pronounced in the first half of the pressure holding process (times 1 to 2) than in the second half (times 3 to 4). They also discovered that if the time integral of the pressure in the first half is taken as the first evaluation value and the time integral of the pressure in the second half is taken as the second evaluation value, the relationship between the first evaluation value and the second evaluation value differs between when temporary clogging occurs and when the system is normal. In other words, they discovered that by focusing on this difference, temporary clogging in some gates can be detected.
[0009] Therefore, the abnormality detection system according to the present invention obtains detection information for detecting a clogged gate based on the first evaluation value and the second evaluation value, which makes it possible to detect a temporary clogged gate that could not be detected in the past, and more reliably detect the clogged gate.
[0010] (2) Preferably, the abnormality detection unit acquires the detection information based on a ratio between the first evaluation value and the second evaluation value. The ratio between the first evaluation value and the second evaluation value is less susceptible to influences such as variations in molding materials and can more accurately represent the influence of temporary clogging.
[0011] (3) Preferably, the abnormality detection unit acquires the detection information indicating that the gate is clogged when the ratio acquired during molding of the molded product to be detected falls outside a predetermined range defined based on the ratio acquired when none of the gates is clogged. With this configuration, temporary clogging of the gate can be easily detected by comparing the first evaluation value and the second evaluation value with the predetermined range.
[0012] (4) Preferably, the abnormality detection unit acquires the detection information by inputting the first evaluation value and the second evaluation value to a trained model that has been machine-learned to learn the correlation between the first evaluation value and the second evaluation value and the clogging of the gate. With this configuration, it is possible to more accurately detect the clogging of the gate even when there is variation in molding conditions.
[0013] (5) Preferably, the first time point is a time point of the pressure holding operation or a time point at which the pressure rise is detected, the fourth time point is a time point of the pressure holding release operation, the second time point is a time point before the middle between the first time point and the fourth time point, and the third time point is a time point after the middle. By configuring in this way, it is possible to more reliably separate an area where the influence of the pressure rise is large from an area where the influence is small.
[0014] (6) Preferably, the pressure sensor is provided in the cold slug well. With this configuration, the pressure can be detected while preventing the pressure sensor from affecting the flow of molding material from the spool to the gate.
[0015] (7) An abnormality detection device according to the present invention is an abnormality detection device that detects an abnormality in a molding device that molds a molded product, the molding device comprising a mold section having therein a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates, an injection section that performs a filling operation of filling the cavity with molten molding material from the spool, a pressure holding operation of holding the pressure of the molding material filled in the cavity, and a pressure holding release operation of releasing the pressure holding of the molding material, and and a pressure sensor that detects the pressure of the molding material in the filling operation, the abnormality detection device comprising: a data acquisition unit that acquires a first evaluation value obtained by integrating over time the pressure detected by the pressure sensor from a first point in time after the filling operation to a second point in time before the pressure-holding release operation, and a second evaluation value obtained by integrating over time the pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the pressure-holding release operation, and an abnormality detection unit that acquires detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value.
[0016] The abnormality detection device according to the present invention obtains detection information for detecting a clogged gate based on the first evaluation value and the second evaluation value, thereby making it possible to detect temporary clogged gates that could not be detected in the past, and to more reliably detect clogged gates.
[0017] (8) An abnormality detection method according to the present invention is an abnormality detection method for detecting an abnormality in a molding device including a mold section having a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates, and an injection section which performs a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, the method comprising the steps of: The abnormality detection method includes a data acquisition process for acquiring a first evaluation value obtained by integrating over time a pressure detected by a pressure sensor that detects the pressure of the molding material in the spool, the runner, or the gate up to a second point in time prior to the pressure release operation, and a second evaluation value obtained by integrating over time a pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the pressure holding release operation, and an abnormality detection process for acquiring detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value.
[0018] The abnormality detection method according to the present invention obtains detection information for detecting a clogged gate based on the first evaluation value and the second evaluation value, thereby making it possible to detect a temporary clogged gate that could not be detected in the past, and to more reliably detect a clogged gate.
[0019] (9) A program according to the present invention is a program for detecting an abnormality in a molding apparatus including a mold section having a spool, a plurality of runners branching off from the spool, a plurality of gates each connected to the plurality of runners, and a cavity connected to the plurality of gates, and an injection section performing a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, the program comprising: a first time point after the filling operation, the program causing a computer device to execute a data acquiring process to acquire a first evaluation value obtained by integrating over time a pressure detected by a pressure sensor detecting the pressure of the molding material in the spool, the runner, or the gate from a time point before the pressure-holding release operation to a second time point before the pressure-holding release operation, and a second evaluation value obtained by integrating over time a pressure detected by the pressure sensor from a third time point after the first time point to a fourth time point after the second time point and before the pressure-holding release operation, and an abnormality detection process to acquire detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value.
[0020] By executing the program according to the present invention, it is possible to obtain detection information for detecting a clogged gate based on the first evaluation value and the second evaluation value, which makes it possible to detect temporary clogged gates that could not be detected in the past, and thus to more reliably detect clogged gates.
[0021] (10) A trained model according to the present invention is a trained model for detecting abnormalities in a molding device having a mold section formed therein with a spool, a plurality of runners branching off from the spool, a plurality of gates each connected to the plurality of runners, and a cavity connected to the plurality of gates, and an injection section which performs a filling operation of filling molten molding material from the spool into the cavity, a pressure holding operation of maintaining the pressure of the molding material filled in the cavity, and a pressure holding release operation of releasing the maintenance of the pressure of the molding material, wherein the explanatory variables include a first evaluation value obtained by integrating over time the pressure detected by a pressure sensor which detects the pressure of the molding material at the spool, the runners, or the gate from a first point in time after the filling operation to a second point in time before the pressure holding release operation, and a second evaluation value obtained by integrating over time the pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the pressure holding release operation, and the objective variable includes a value related to clogging of the gate.
[0022] By inputting the first evaluation value and the second evaluation value into the trained model according to the present invention, it is possible to obtain detection information for detecting a clogged gate. This makes it possible to detect temporary clogged gates that could not be detected in the past, and therefore to more reliably detect clogged gates. Effect of the Invention
[0023] According to the present invention, a blockage of the gate can be detected more reliably. [Brief description of the drawings]
[0024] [Figure 1] 1 is a block diagram illustrating a schematic diagram of an anomaly detection system according to a first embodiment. [Diagram 2] FIG. 2 is an explanatory diagram conceptually showing the molding device in FIG. [Diagram 3] FIG. 2 is an explanatory diagram conceptually showing the molding device in FIG. [Figure 4] FIG. 4 is a cross-sectional view showing the mold part of FIG. [Diagram 5] 5 is a cross-sectional view showing a mold part cut along the cutting line of arrow V in FIG. 4. [Figure 6] 1 is an example of a graph showing time series data of pressure. [Figure 7] FIG. 13 is a diagram illustrating gate clogging. [Figure 8] FIG. 4 is a diagram illustrating a first evaluation value and a second evaluation value. [Figure 9] 1 is a block diagram showing a functional configuration of a learning device according to a first embodiment. [Figure 10] 1 is a block diagram showing a functional configuration of an abnormality detection device according to a first embodiment. [Figure 11] FIG. 11 is a block diagram illustrating a schematic diagram of an anomaly detection system according to a second embodiment. [Figure 12] FIG. 11 is a block diagram showing the functional configuration of an abnormality detection device according to a second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0025] First Embodiment A first embodiment of the present invention will be described below with reference to the drawings.
[0026] <Overall configuration of anomaly detection system> 1 is a block diagram illustrating a schematic diagram of an abnormality detection system 10 according to a first embodiment. The abnormality detection system 10 includes a plurality of forming devices 20, a learning device 30, an abnormality detection device 40, an input unit 50, and a display unit 60.
[0027] The molding device 20, the learning device 30, the anomaly detection device 40, the input unit 50, and the display unit 60 are each provided to be capable of communicating wirelessly or via wires. The learning device 30 and the anomaly detection device 40 are configured by an information processing device (computer device) having a calculation unit (e.g., CPU, GPU, etc.) and a storage unit (e.g., HDD, SSD, etc.). The learning device 30 and the anomaly detection device 40 may be configured by the same information processing device, or may be configured by separate information processing devices.
[0028] In this embodiment, a plurality of forming apparatuses 20 are connected to one learning device 30 and one abnormality detection device 40, and the learning device 30 and the abnormality detection device 40 perform learning and abnormality detection based on various data transmitted from the plurality of forming apparatuses 20. Note that the forming apparatuses 20 may have a one-to-one correspondence with the learning device 30 and the abnormality detection device 40. That is, the abnormality detection system 10 may include one forming apparatus 20, or may include a plurality of learning devices 30 and abnormality detection devices 40.
[0029] The input unit 50 is, for example, a keyboard or a mouse, and accepts various inputs from an operator. The display unit 60 is, for example, a display or a speaker, and displays various information in the anomaly detection system 10. The input unit 50 and the display unit 60 may be integrated together, for example, as a touch panel. Furthermore, the input unit 50 and the display unit 60 may be provided as a portable terminal device that can be moved to a location away from the molding device 20, the learning device 30, and the anomaly detection device 40.
[0030] <Outline of molding device configuration> 2 and 3 are explanatory diagrams conceptually showing molding apparatus 20. In Fig. 2 and Fig. 3, the parts shown as cross sections are hatched. Molding apparatus 20 has a bed 21, an injection section 22, a mold clamping section 23, a mold section 24, a pressure sensor 25, a temperature sensor 26, and a control panel 27. Fig. 2 shows molding apparatus 20 in a state where mold section 24 is open, and Fig. 3 shows molding apparatus 20 in a state where mold section 24 is assembled. Molding apparatus 20 is an apparatus that performs mold clamping type injection molding.
[0031] The control panel 27 has a control unit 271 and a communication unit 272. The control unit 271 is electrically connected to each drive unit (motor 237, etc.) of the molding device 20, and outputs operation commands to each drive unit. The control unit 271 is also electrically connected to each sensor (pressure sensor 25, etc.) of the molding device 20, and signals detected by each sensor are input to the control unit 271. The control unit 271 is configured by an information processing device having a calculation unit (e.g., CPU, GPU, etc.) and a storage unit (e.g., HDD, SSD, etc.).
[0032] The communication unit 272 communicates with other units (such as the learning device 30) of the anomaly detection system 10. For example, the communication unit 272 transmits signals detected by the sensors to the learning device 30 or the anomaly detection device 40. The communication unit 272 also receives determination information and prediction information, which will be described later, from the anomaly detection device 40.
[0033] The mold clamping unit 23 has a fixed platen 231, a movable platen 232, tie bars 233, a ball screw 234, a support platen 235, a mold clamping force sensor 236, and a motor 237. The fixed platen 231 and the support platen 235 are fixed to the bed 21. The support platen 235 supports a ball screw 234. The ball screw 234 is connected to a motor 237. When the motor 237 is rotated by an operation command from the control unit 271, the ball screw 234 moves. The movable platen 232 is fixed to the end of the ball screw 234 opposite to the end connected to the motor 237.
[0034] Here, in the molding apparatus 20, the direction in which the ball screw 234 moves is referred to as the "axial direction." The side on which the motor 237 is located relative to the ball screw 234 is referred to as "one side" in the axial direction, and the side on which the movable platen 232 is located relative to the ball screw 234 is referred to as "the other side" in the axial direction.
[0035] The movable platen 232 moves in the axial direction in accordance with the movement of the ball screw 234. A through hole 232a penetrating in the axial direction is formed in the movable platen 232. One axial end of the tie bar 233 is fixed to the support platen 235, and the other axial end is fixed to the fixed platen 231. The tie bar 233 is inserted into the through hole 232a of the movable platen 232. In this way, the tie bar 233 guides the axial movement of the movable platen 232.
[0036] The mold clamping force sensor 236 detects pressure (reaction force of the mold clamping force) applied to the support platen 235 from the ball screw 234. The mold clamping force sensor 236 outputs a detection signal related to the pressure to the control unit 271. The mold clamping force sensor 236 may be installed at another position as long as it is a position where the mold clamping force in the mold unit 24 described later can be detected. The fixed platen 231 is formed with a through hole 231a whose diameter expands on the other axial side. A cylinder 222 described later is inserted into the through hole 231a.
[0037] The mold section 24 has a plurality of molds 241, 242. The mold 241 is fixed to a movable platen 232. When the ball screw 234 moves in the axial direction, the mold 241 also moves in the axial direction together with the movable platen 232. That is, the mold 241 is a movable mold. The mold 242 is fixed to the fixed platen 231. That is, the mold 242 is a fixed mold.
[0038] Please refer to Fig. 3. When the mold 241 moves to the other axial direction by the ball screw 234 and comes into contact with the mold 242 (i.e., when the multiple molds 241, 242 are combined), a cavity C1 is formed between the molds 241, 242. The cavity C1 in this embodiment is an annular space into which the molding material is filled.
[0039] Fig. 4 is an enlarged cross-sectional view of the mold part 24 shown in Fig. 3. A spool 243, a plurality of runners 244 (four in this embodiment) branching off from the spool 243, and a plurality of gates 245 (four in this embodiment) respectively connected to the plurality of runners 244 are formed inside the mold 242. A cavity C1 is connected to the plurality of gates 245.
[0040] The spool 243 is a passage into which the molding material injected from the injection section 22 into the mold section 24 first flows. An end 243a of the spool 243 on the injection section 22 side opens so as to communicate with a nozzle 224 described below. The inner diameter of the spool 243 is smallest at the end 243a, and increases from the end 243a to the runner 244.
[0041] The multiple runners 244 branch off slightly closer to the end 243a than the end 243b on the cavity C1 side (i.e., the side opposite the end 243a) of the spool 243. The space on the end 243b side of the position where the runners 244 branch off in the spool 243 is a resin reservoir for storing cold slug, and is also called a "cold slug well."
[0042] FIG. 5 is a cross-sectional view showing the mold part 24 taken along the cutting line indicated by the arrow V in FIG. The gates 245 correspond one-to-one to the runners 244. The inner diameter of the gates 245 decreases toward the cavity C1. Hereinafter, when the gates 245 are to be specifically distinguished from one another, a specific gate 245 (e.g., the gate 245 located at the top of FIG. 5) will be referred to as gate GT1, and the remaining three gates 245 in the clockwise direction from the gate GT1 will be referred to as gates GT2, GT3, and GT4, respectively.
[0043] Pressure sensor 25 is provided at end 243b of spool 243, and detects the pressure of the molding material (in a molten state, a solidified state, or a mixture of a molten state and a solidified state) in spool 243. More specifically, pressure sensor 25 is provided in a cold slug well. By providing pressure sensor 25 at this position, it is possible to detect the pressure while preventing pressure sensor 25 from affecting the flow of molding material from spool 243 to gate 245.
[0044] The position of the pressure sensor 25 is not limited to this. For example, the pressure sensor 25 may be provided at another position on the spool 243, or may be provided on any of the multiple runners 244 and the multiple gates 245. That is, the pressure sensor 25 detects the pressure of the molding material in the spool 243, the runner 244, or the gate 245. In other words, the pressure sensor 25 detects the pressure of the molding material downstream of a nozzle 224 described below and upstream of a cavity C1. The pressure sensor 25 outputs a detection signal related to the pressure to the control unit 271.
[0045] The number of pressure sensors 25 is not limited. In this embodiment, an example in which there is one pressure sensor 25 will be described, but there may be two or more pressure sensors 25. For example, one pressure sensor 25 may be provided for each runner 244, resulting in a total of four pressure sensors 25. In this case, the time series data of pressure described later may be acquired as time series data of the average value of the pressures detected by the four pressure sensors 25.
[0046] 2. The injection section 22 has a hopper 221, a cylinder 222, a screw 223, a ball screw 225, a motor 226, a pressure sensor 227, a movement amount sensor 228, and a heater 229. The hopper 221 is connected to the cylinder 222 and supplies molding material into the cylinder 222. The cylinder 222 is a member having a hollow cylindrical shape extending in the axial direction. The diameter of the end on one axial side of the cylinder 222 narrows as it approaches the extreme end on one radial side, and a nozzle 224 is provided at the extreme end on one axial side of the cylinder 222. The nozzle 224 is connected to a spool 243 of a mold 242.
[0047] The screw 223 is inserted into the cylinder 222 from the other axial end of the cylinder 222. A ball screw 225 is connected to the other axial end of the screw 223, and a motor 226 is connected to the other axial end of the ball screw 225. When the motor 226 rotates in response to an operation command from the control unit 271, the ball screw 225 moves in the axial direction. Accordingly, the screw 223 also moves in the axial direction. At this time, the screw 223 rotates in a circumferential direction with the axial direction as the central axis.
[0048] The pressure sensor 227 detects the pressure (reaction force of the pushing force of the screw 223) applied from the ball screw 225 to the motor 226. That is, the pressure sensor 227 detects the pressure that the screw 223 receives from the molding material. The pressure sensor 227 outputs a detection signal related to the pressure to the control unit 271. Note that the pressure sensor 227 may be installed in another position as long as it is a position where the pushing force of the screw 223 can be detected.
[0049] The movement amount sensor 228 detects the amount of movement in the axial direction of the ball screw 225. The movement amount sensor 228 outputs a detection signal related to the amount of movement to the control unit 271. Note that the movement amount sensor 228 may be installed in another position as long as the movement amount in the axial direction of the ball screw 225 can be detected.
[0050] The heater 229 is, for example, a resistance heater formed by winding a resistance wire into a coil shape. When a current flows through the resistance wire in response to an operation command from the control unit 271, the heater 229 heats the inside of the cylinder 222 by resistance heat.
[0051] The temperature sensor 26 is built into the mold 242 and detects the temperature of the mold 242. The temperature sensor 26 outputs a detection signal related to the temperature to the control unit 271. The temperature sensor 26 may be installed in the mold 241. The temperature sensor 26 may also be installed in the cylinder 222. That is, it is only necessary for the temperature sensor 26 to be able to directly or indirectly detect the temperature of the molding material supplied into the cavity C1.
[0052] <Manufacturing method using molding equipment> 2 and 3, a method for manufacturing a molded product using the molding apparatus 20 will be described. In the method for manufacturing a molded product using the molding apparatus 20, a pre-process ST1, a mold clamping process ST2, a filling process ST3, a pressure holding process ST4, a pressure releasing process ST5, and a demolding process ST6 are performed in this order. In this embodiment, the molded product is a resin cage used for a rolling bearing. However, this is only one example of a molded product, and the molded product molded by the molding apparatus according to the present invention may be a molded product of other shapes and applications.
[0053] Please refer to Fig. 2. First, a pre-process ST1 is performed. In the pre-process ST1, the screw 223 is rotated by the motor 226, and pellets of the molding material are supplied from the hopper 221 into the cylinder 222 while the inside of the cylinder 222 is heated by the heater 229. The pellets of the molding material are melted in the cylinder 222 by frictional heat caused by the rotation of the screw 223 and heating by the heater 229, and become a molten molding material. When a predetermined amount of molding material is stored in the cylinder 222, the pre-process ST1 is completed.
[0054] Next, when the mold clamping step ST2 is started, in the molding apparatus 20 in the state shown in FIG. 2, the ball screw 234 moves to the other axial side in response to an operation command from the control unit 271, and the mold 241 is brought into contact with the mold 242 as shown in FIG. 3. With the mold 241 and the mold 242 combined in this manner, the ball screw 234 further presses the mold 241 against the mold 242 in the other axial direction with a predetermined mold clamping force. That is, the multiple molds 241, 242 are clamped together. As a result, an annular cavity C1 is formed between the multiple molds 241, 242. With the above, the mold clamping step ST2 is completed.
[0055] Here, the mold clamping force is one of the molding conditions, and is determined according to other molding conditions such as the shapes of the molds 241 and 242. The mold clamping force is detected by a mold clamping force sensor 236.
[0056] Next, when the filling step ST3 is started, the ball screw 225 moves to one axial side while maintaining the above-mentioned mold clamping force. This causes the screw 223 to push the molding material to one axial side, and the molten molding material is injected from the nozzle 224 of the cylinder 222 through the spool 243, the multiple runners 244, and the multiple gates 245 into the cavity C1 (filling operation).
[0057] When the molding material is filled into the cavity C1, the filling step ST3 is completed. In the filling step ST3, the molten molding material is supplied to the cavity C1 from the vicinity of the surfaces of the metal molds 241 and 242 while gradually solidifying.
[0058] Next, when the pressure holding step ST4 is started, the screw 223 further pushes the molding material toward one side in the axial direction, and the molding material is further injected from the nozzle 224 of the cylinder 222 into the cavity C1. As a result, a predetermined pressure (e.g., several tens to several hundreds of MPa) is applied to the molding material filled in the cavity C1. Then, the screw 223 maintains this state for a predetermined time, thereby continuing to apply the predetermined pressure to the molding material for a predetermined time (e.g., several seconds) (pressure holding operation). The pressure (pressure) with which the screw 223 pushes out the molding material into the cavity C1 is detected by the pressure sensor 227.
[0059] Next, when the pressure-removing step ST5 is started, the screw 223 moves to the other axial side and releases the pressure of the molding material (pressure-removing operation). After the pressure-removing operation, a predetermined time has elapsed and the pressure of the molding material in the cavity C1 falls below a predetermined value, and the pressure-removing step ST5 ends. Thereafter, when the mold release step ST6 is started, the mold part 24 is cooled, so that the molding material in the cavity C1 solidifies and a molded product is formed. Then, the ball screw 234 moves to one axial side, and the mold 241 is separated from the mold 242, and the molded product is removed. Note that cooling of the mold part 24 may be started simultaneously with the pressure-removing step ST5.
[0060] <About gate clogging> Next, the clogging of the gate 245 detected in this embodiment will be described. 6 is an example of a graph showing time series data of pressure detected by each sensor of molding apparatus 20 in filling step ST3, pressure holding step ST4, and pressure releasing step ST5. In Fig. 6, the vertical axis represents pressure P, and the horizontal axis represents time t.
[0061] In this embodiment, the origin of time (t=0) is the time when the control unit 271 issues an operation command to the motor 226 to move the screw 223 to the other axial side. In other words, it is the time when the injection unit 22 starts to supply the molding material to the mold unit 24 in the filling process ST3. Note that the origin of time (t=0) is not limited to this, and may be the time when the pressurization sensor 227 detects a predetermined pressure, or the time when the movement amount sensor 228 detects a predetermined movement amount.
[0062] Graph line F0 in FIG. 6 is time-series data of the pressure (pressurization) detected by pressurization sensor 227 when no blockage occurs in any of the multiple gates 245 (i.e., normal). The pressure of graph line F0 rises to pressure Ps1 in the filling step ST3, and at time X1 when pressure Ps1 is reached, filling step ST3 ends. Here, pressure Ps1 is the "set pressure" for pressure holding step ST4. That is, during pressure holding step ST4, control unit 271 feedback-controls motor 226 so that the pressure detected by pressurization sensor 227 is maintained at Ps1. Therefore, during pressure holding step ST4, the pressure detected by pressurization sensor 227 is approximately constant at Ps1.
[0063] The pressure holding step ST4 starts at time X1 and ends at time X4. That is, the pressure holding time is (X4-X1). The pressure holding time is set in advance in the control unit 271. Thereafter, the pressure is reduced in the pressure holding release step ST5. The pressure holding release step ST5 starts at time X4. That is, the pressure holding release operation is executed at time X4.
[0064] 6 is time-series data of the pressure detected by the pressure sensor 25 under normal conditions. The graph line F1 rises after the graph line F0 rises in the filling step ST3, the pressure rises to a peak in the pressure holding step ST4, and the pressure suddenly drops in the pressure holding release step ST5.
[0065] There are three patterns of clogging of the gates 245, and any of the clogging patterns may cause abnormalities in the molded product. The first pattern is a pattern in which all the gates 245 are clogged. Hereinafter, this pattern is appropriately referred to as "all clogging." This pattern also includes cases in which all the runners 244 are clogged and cases in which the spool 243 is clogged. The graph line F0a is time-series data of the pressure detected by the pressurization sensor 227 when all the gates 245 are clogged. When all the gates 245 are clogged, the space in the mold section 24 into which the molding material can be poured becomes narrow, so that the space is filled with a small injection amount. As a result, as shown by the graph line F0a, the pressure detected by the pressurization sensor 227 reaches pressure Ps1 at time X1a, which is earlier than time X1 in the normal state.
[0066] Therefore, if the filling time (from time 0 to time X1a) detected by the pressurization sensor 227 is shorter than the normal filling time, there is a high possibility that all of the gates 245 are clogged. In other words, by focusing on the length of the filling time detected by the pressurization sensor 227, it is possible to detect that all of the gates 245 are clogged. If all of the gates 245 are clogged, the molding material does not flow into the cavity C1, and a molded product is not molded in the first place. Therefore, it is possible to detect that all of the gates 245 are clogged by observing with a sensor such as a camera that a molded product is not molded in the demolding step ST6.
[0067] The second clogging pattern is a pattern in which some of the gates 245 are clogged. Hereinafter, this pattern will be referred to as "partial clogging". For example, the gate GT1 shown in FIG. 5 is clogged, and the other gates GT2, GT3, and GT4 are not clogged. In this case, the molding material flows into the cavity C1 from the other gates GT2, GT3, and GT4 that are not clogged, so the pressure detected by the pressurizing sensor 227 is almost the same as the pressure in the normal state. For this reason, clogging of some of the gates 245 cannot be detected based on the pressure of the pressurizing sensor 227. In addition, since a molded product is also molded, clogging of some of the gates 245 cannot be detected even if a camera is used to observe whether a molded product is molded in the demolding process ST6.
[0068] When such clogging occurs, molding material does not flow into the cavity C1 near the clogged gate 245 as usual, which has the adverse effect of increasing the variation in the dimensions, weight, and quality (e.g., strength) of the molded product. In addition, weld areas may be formed in unintended areas (areas other than the designed areas), which may reduce the strength of the molded product. For this reason, it is important to have a technology that can sense the clogging of some of the gates 245 in the molding device 20 and automatically determine it.
[0069] Therefore, in this embodiment, the pressure sensor 25 detects the pressure of the molding material downstream of the nozzle 224 and upstream of the cavity C1, and detects clogging of some of the gates 245 based on the detected pressure. The graph line F2 is time-series data of the pressure detected by the pressure sensor 25 when some of the gates 245 are clogged. When some of the gates 245 are clogged, the molding material flowing from the nozzle 224 to the spool 243 has less space to escape, so that the pressure inside the spool 243, the multiple runners 244, and the multiple gates 245 increases more than normal. As a result, as shown by the graph line F2, the pressure detected by the pressure sensor 25 in the pressure holding step ST4 becomes higher than the pressure of the graph line F1 in the normal state.
[0070] For this reason, if the pressure during the pressure holding step ST4 is higher than the pressure during the pressure holding step ST4 under normal conditions, there is a high possibility that some of the gates 245 are clogged. In other words, by focusing on the level of the pressure during the pressure holding step ST4 detected by the pressure sensor 25, it is possible to detect that some of the gates 245 are clogged. Here, due to noise or the like of the pressure sensor 25, the pressure of the graph line F1 may be higher than the pressure of the graph line F2 at only one predetermined point in time during the pressure holding step ST4. In this way, if attention is focused only on the point in time affected by noise, partial clogging cannot be detected correctly.
[0071] Therefore, in this embodiment, the influence of noise, etc. is reduced by comparing the average or time integral of the pressure detected by the pressure sensor 25 during a predetermined period during the pressure holding process ST4 (for example, from time X1 to time X4) with the average or time integral of the pressure during the same period under normal conditions.
[0072] The third clogging pattern is a pattern in which some of the gates 245 are clogged, but the clogging of the gates 245 is cleared during the pressure retention step ST4. Hereinafter, this pattern will be appropriately referred to as "temporary clogging."
[0073] FIG. 7 is a diagram for explaining clogging of the gate 245 due to this pattern. FIG. 7(a) and FIG. 7(b) show the same cross section as FIG. 5, and show the state of the four gates GT1 to GT4 at time points X1 and X4, respectively. As shown in FIG. 7(a), at time point X1 (start time of pressure holding step ST4), a foreign object Fm1 clogs the gate GT1, and the gate GT1 is blocked. The other gates GT2, GT3, and GT4 are not clogged, and the molding material flows into the cavity C1 from the gates GT2, GT3, and GT4. The state up to this point is the same as the second pattern described above. That is, since there is less room for the molding material flowing from the nozzle 224 into the spool 243 to escape, the pressure inside the spool 243, the multiple runners 244, and the multiple gates 245 rises higher than normal.
[0074] Here, the foreign matter Fm1 that clogs the gate 245 may be, for example, solidified molding material. The temperature inside the cylinder 222 is higher than that of the mold section 24 because it is heated by the heater 229. Therefore, the molding material that is first injected from the nozzle 224 in the filling step ST3 may be partially solidified in the spool 243 or the multiple runners 244 by being cooled by the mold 242, and may become cold slugs.
[0075] If the foreign object Fm1 is a solidified molding material (cold slug), the foreign object Fm1 may be broken, melted, or released from its position on the inner wall of the gate GT1 due to the pressure rise inside the gate GT1 during the pressure holding step ST4 (before the time point X4), causing the foreign object Fm1 to flow into the cavity C1. This clears the blockage of the gate GT1, as shown in FIG. 7(b).
[0076] Graph line F3 is time series data of the pressure detected by pressure sensor 25 when temporary clogging occurs in gate 245. The clogging of gate 245 is cleared, for example, before time X2. Because some of gates 245 are clogged at time X1, the pressure detected in the first half of pressure holding process ST4 (from time X1 to time X2) is higher than the pressure of graph line F1 under normal conditions. And, because the clogging is cleared at some point before time X2, the pressure detected in the second half of pressure holding process ST4 (from time X2 to time X4) is almost the same as the pressure of graph line F1 under normal conditions.
[0077] In this way, when some of the gates 245 are temporarily clogged and then the clog is cleared, even if attention is focused on the pressure during the pressure holding step ST4, there is almost no difference from the pressure during the pressure holding step ST4 under normal conditions depending on the time point of comparison, so it is not possible to accurately detect the clog of the gate 245. Also, when comparing the average or time integral of the pressure over the entire period of the pressure holding step ST4 (from time point X1 to time point X4), as shown by graph line F3, the pressure increase occurs unevenly in the first half of the pressure holding step ST4, so when the pressure over that entire period is averaged or integrated over time, there is almost no difference from the normal state, and the temporary clog of the gate 245 cannot be accurately detected.
[0078] Furthermore, if temporary clogging occurs, there is a risk that the molded product will have a larger variation in size, weight, and quality, or that its strength will decrease, as in the case of the second pattern. Therefore, the inventors came up with the idea of dividing the pressure in the dwelling process ST4 into a first half portion where the influence of the pressure increase is large, and a second half portion where the influence of the pressure increase is small, and detecting temporary clogging of the gate 245 based on the time integral of the pressure in the first half portion and the time integral of the pressure in the second half portion.
[0079] <About the first and second evaluation values> FIG. 8 is a diagram for explaining the first evaluation value A1 and the second evaluation value A2 used in the abnormality detection of this embodiment. FIG. 8(a) is a graph showing the relationship between the graph line F3 of FIG. 6 and the first evaluation value A1 and the second evaluation value A2. In the graph of FIG. 8(a), the vertical axis represents pressure, and the horizontal axis represents time. FIG. 8(b) is a graph for explaining a method for detecting an abnormality based on the first evaluation value A1 and the second evaluation value A2. In the graph of FIG. 8(b), the vertical axis represents the second evaluation value A2, and the horizontal axis represents the first evaluation value A1.
[0080] The first evaluation value A1 is a value obtained by integrating the pressure detected by the pressure sensor 25 over time from a first time point (time point X1) after the filling operation to a second time point (time point X2) before the pressure-holding release operation. As shown in FIG. 8(a), the first evaluation value A1 corresponds to the area of the graph line F3 from time point X1 to time point X2. Here, the graph line F3 is actually composed of a plurality of plots. Therefore, the first evaluation value A1 is calculated as the sum of the plots of the graph line F3 from time point X1 to time point X2 in terms of data. Note that the plurality of plots constituting the graph line F3 may be original data (raw data acquired from the pressure sensor 25) or may be a moving average of the original data (for example, a two-point moving average).
[0081] The second evaluation value A2 is a value obtained by integrating the pressure detected by the pressure sensor 25 over time from a third time point (time point X3) that is later than the first time point (time point X1) to a fourth time point (time point X4) that is later than the second time point (time point X2) and before the pressure-retaining release operation. As shown in FIG. 8(a), the second evaluation value A2 corresponds to the area of the graph line F3 from time point X3 to time point X4. As with the first evaluation value A1, the second evaluation value A2 is calculated as the sum of the plots of the graph line F3 from time point X3 to time point X4 in terms of data.
[0082] 8(a), the first, second, third, and fourth time points are respectively referred to as time points X1, X2, X3, and X4. Time point X1 is the start time of the pressure maintaining process ST4, and is the time point when the pressure detected by the pressure sensor 227 reaches the set pressure Ps1 of the pressure maintaining process.
[0083] The first time point is not limited to time point X1. For example, the first time point may be the time point when the rising edge of the graph line F3 is detected. In this case, the first time point is the time point when the pressure of the graph line F3 exceeds a predetermined percentage (e.g., 50%) of the maximum pressure in the pressure holding step ST4. The first time point may also be the start of the filling step (i.e., t=0).
[0084] Time point X2 is a predetermined time t1 after time point X1 (X2=X1+t1). Time point X3 is a predetermined time t2 before time point X4 (X3=X4-t2). The predetermined times t1 and t2 are each shorter than the time from time point X1 to time point X4 (X4-X1). Therefore, at least a part of the first half region (X1-X2) for calculating the first evaluation value A1 is located before the second half region (X3-X4) for calculating the second evaluation value A2. At least a part of the second half region is located after the first half region.
[0085] In this embodiment, the predetermined time t1 and the predetermined time t2 are each equal to or less than half the time (X4-X1) from time X1 to time X4. Therefore, time X2 is a time before the halfway point between time X1 and time X4. Furthermore, time X3 is a time after the halfway point between time X1 and time X4. A temporary clogging of the gate 245 is difficult to detect in the same manner as a partial clogging of the gate 245, especially when the clogging is cleared before the halfway point between time X1 and time X4, because the effect of the pressure increase is biased toward the front half.
[0086] In this embodiment, by setting time X2 to a time before the midpoint between time X1 and time X4, and setting time X3 to a time after the midpoint, it is possible to more reliably separate areas where the influence of the pressure rise is large from areas where the influence is small.
[0087] In FIG. 8(a), the time point X2 is a time point earlier than the time point X3, but is not limited to this. The time point X2 may be the same as the time point X3. Also, the time point X2 may be a time point later than the time point X3. In this case, the first half region and the second half region partially overlap.
[0088] Time point X4 is the end time of the pressure holding process ST4, and is calculated, for example, by adding a pressure holding time previously set in the control unit 27 to time point X1. The fourth time point is not limited to time point X4, and may be a time point prior to time point X4. However, the fourth time point is later than time points X2 and X3.
[0089] In this way, the first evaluation value A1 and the second evaluation value A2 are calculated based on the time series data of pressure acquired from the pressure sensor 25, and then the first evaluation value A1 and the second evaluation value A2 are plotted as shown in Fig. 8(b). When multiple pieces of data from normal times when there is no blockage in the gate 245 are plotted, a scatter diagram from normal times is drawn. Then, a regression line L1 (i.e., A2 = b x A1) passing through the origin, an upper limit line Lmax (i.e., A2 = b x A1 + α1), and a lower limit line Lmin (i.e., A2 = b x A1 - α1) are drawn for the scatter diagram.
[0090] Here, b corresponds to the average value (or median) of the ratio between the first evaluation value A1 and the second evaluation value A2 under normal conditions. The pressure value detected by the pressure sensor 25 varies due to variations in the molding material, etc., so the first evaluation value A1 and the second evaluation value A2 also take various values as shown in FIG. 8(b). However, the ratio between the first evaluation value A1 and the second evaluation value A2 (i.e., the relative relationship) is almost the same value. For this reason, the comparison is made based on the ratio between the first evaluation value A1 and the second evaluation value A2, which has less variation.
[0091] Moreover, the intercept ±α1 is a value that is appropriately set based on, for example, the distribution of the ratio between the first evaluation value A1 and the second evaluation value A2. Therefore, when the first evaluation value A1 and the second evaluation value A2 calculated during normal times are newly plotted on the graph in Fig. 8(b) (for example, point P1), they are plotted between the upper limit line Lmax and the lower limit line Lmin. In other words, if the calculated first evaluation value A1 and the second evaluation value A2 are plotted on the graph in Fig. 8(b) and fall between the upper limit line Lmax and the lower limit line Lmin, it can be said that no temporary clogging occurred in the gate 245.
[0092] Furthermore, when the first evaluation value A1 and the second evaluation value A2 calculated when a temporary blockage occurs in the gate 245 are newly plotted on the graph in Fig. 8(b) (for example, point P2), they are plotted below the lower limit line Lmin. In other words, if the calculated first evaluation value A1 and second evaluation value A2 are plotted below the lower limit line Lmin when plotted on the graph in Fig. 8(b), it can be said that a temporary blockage occurred in the gate 245.
[0093] As described above, the inventors have found, through extensive research, that when temporary clogging occurs in the gate 245, the pressure sensor 25 detects pressure, and the first half of the pressure holding step ST4 is more affected by the pressure rise than the second half. The inventors have found that when the time integral of the pressure in the first half is quantified using the first evaluation value A1, the time integral of the pressure in the second half is quantified using the second evaluation value A2, and the ratio (A2 / A1) of the first evaluation value A1 and the second evaluation value A2 is calculated, the ratio decreases by the amount of the pressure rise in the first half when temporary clogging occurs. Based on this, the inventors have come up with an invention for detecting temporary clogging of the gate 245 based on whether the ratio decreases below a predetermined value.
[0094] Therefore, in the abnormality detection system 10 according to this embodiment, the learning device 30 generates a trained model Tm1 that learns the correlation between the first evaluation value A1 and the second evaluation value A2 acquired based on the pressure of the pressure sensor 25 under normal conditions and the clogging of the gate 245, and the abnormality detection device 40 acquires detection information for detecting clogging of the gate 245 based on the trained model Tm1, the first evaluation value A1, and the second evaluation value A2. The learning device 30 and the abnormality detection device 40 will be described below.
[0095] <Explanation of the learning device> 9 is a block diagram showing the functional configuration of the learning device 30 according to this embodiment. The learning device 30 has a training data acquisition unit 31, a learning calculation unit 32, a molding information storage unit 33, and a learned model storage unit 34. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as a HDD.
[0096] Various molding information is stored in the molding information storage unit 33. The molding information is, for example, information in a table format in which various first information and second information are associated with each other. For example, when the first information is the type of mold, the second information includes various dimensions of the mold and the volume of the cavity C1. When the first information is the type or lot number of the molding material, the second information includes the physical properties of the molding material (viscosity, moisture content, etc.).
[0097] The training data acquisition unit 31 acquires information about training data from each part of the anomaly detection system 10. The training data includes gate information, molding information corresponding to the gate information, time point X1, first evaluation value A1, and second evaluation value A2. The training data is acquired based on data detected by each part of the anomaly detection system 10 (e.g., pressure sensor 25) when a molded product to be learned is molded, for example.
[0098] The gate information is a value related to the blockage of the gate 245, and more specifically, is a value that quantifies the degree of blockage of the gate 245. For example, a complete blockage is quantified as "3," a partial blockage as "2," a temporary blockage as "1," and a state in which the gate 245 is not clogged as "0."
[0099] The training data acquisition unit 31 acquires gate information through input by an operator. For example, for the purpose of learning the learning device 30, the operator artificially clogs the gate 245 and inputs the manner in which the gate 245 is clogged into the input unit 50. In this way, the training data acquisition unit 31 acquires the gate information.
[0100] Furthermore, the training data acquisition unit 31 acquires molding information based on information input by an operator to the input unit 50 and the molding information storage unit 33. For example, when molding a molded product to be learned, the operator inputs the lot number of a molding material related to the molded product. The training data acquisition unit 31 acquires molding information related to the physical properties (e.g., viscosity) of the molding material corresponding to the lot number from the molding information storage unit 33.
[0101] The molding device 20 molds a molded product multiple times using a gate 245 that is not clogged and a gate 245 that is artificially clogged. Then, the training data acquisition unit 31 acquires a time point X1 at which gate information is assigned as teacher data, a first evaluation value A1, and a second evaluation value A2 based on time series data of pressure detected by the pressure sensor 25. For example, "0" is assigned as gate information to the time point X1, the first evaluation value A1, and the second evaluation value A2 acquired when a molded product is molded using the gate 245 that is not clogged.
[0102] The training data acquisition unit 31 further acquires a plurality of environmental values as training data. The environmental values are, for example, values related to temperature detected by the temperature sensor 26 when the molded article to be learned is molded. The environmental values may further include values related to the environment around and inside the molding apparatus 20 detected by other sensors (e.g., humidity sensors) not shown when the molded article to be learned is molded.
[0103] The training data acquisition unit 31 acquires one set of training data (a set of gate information, molding information, time point X1, first evaluation value A1, second evaluation value A2, and environmental value acquired during molding of the molded product) for example, each time a molded product to be learned is molded. The training data acquisition unit 31 acquires multiple sets of training data for the predetermined number of times by molding the molded product to be learned a predetermined number of times.
[0104] The learning calculation unit 32 performs calculations for supervised machine learning based on a plurality of sets of training data to generate a learned model Tm1 that models correlations between the shaping information, the time point X1, the first evaluation value A1, the second evaluation value A2, and the environment value, and the gate information. In this embodiment, a convolutional neural network (CCN) is used as the machine learning model, but other models may be used. For example, a regression tree model that is a model for grouping data may be used.
[0105] Specifically, when generating the trained model Tm1, the forming information, the time point X1, the first evaluation value A1, the second evaluation value A2, and the environmental value (collectively referred to as "input information") are used as explanatory variables, and the gate information is used as a target variable, thereby modeling the correlation between the input information and the gate information. That is, the learning calculation unit 32 generates the trained model Tm1 that constitutes an intermediate layer for outputting the gate information corresponding to the input information from the output layer when the input information is input.
[0106] Here, the tendency of the trained model Tm1 will be described. As shown by the graph line F0a in Fig. 6, the time point X1 tends to be earlier than the normal time point when all the gates 245 are clogged. Therefore, when the value of the time point X1 input to the input layer is smaller than the normal learning value, the output layer has a higher probability of the gate information indicating that all the gates 245 are clogged (in the above example, "3").
[0107] 6, the first evaluation value A1 and the second evaluation value A2 (i.e., the time integral of the pressure during a predetermined period of the pressure holding step ST4) tend to be larger than normal when some of the gates 245 are clogged. For this reason, when both the first evaluation value A1 and the second evaluation value A2 input to the input layer are larger than the normal learning values, the output layer has a higher probability of the gate information indicating that some of the gates 245 are clogged ("2" in the above example).
[0108] 8(b), the ratio (A2 / A1) of the first evaluation value A1 and the second evaluation value A2 tends to be smaller than the normal value when a temporary blockage occurs in the gate 245. For this reason, when the ratio of the first evaluation value A1 and the second evaluation value A2 input to the input layer is smaller than the normal learning value, the output layer has a higher probability of the gate information (in the above example, "1") indicating that a temporary blockage has occurred in the gate 245.
[0109] If the input layer does not have any of the above tendencies and a value close to the normal learning value is input, the output layer will have a high probability of showing that gate information ("0" in the above example) indicating that gate 245 is not clogged.
[0110] When generating the trained model Tm1, the input information only needs to include the first evaluation value A1 and the second evaluation value A2, and does not need to include the time point X1, the shaping information, and the environmental value. However, when the environmental value and the shaping information vary greatly, or when the variation has a large effect on the clogging of the gate 245, it is preferable to include the environmental value and the shaping information in the input information when generating the trained model Tm1 in order to more accurately predict the gate information corresponding to the input information.
[0111] Here, the time point X1 is mainly learned as a value indicating whether or not all of the gates 245 are clogged. Whether or not all of the gates 245 are clogged can be known based on whether or not a molded product is molded in the molding device 20, so if the abnormality detection system 10 is to detect only partial and temporary blockages of the gates 245 and not complete blockages, the time point X1 may be omitted from the input information.
[0112] The trained model Tm1 generated by the learning calculation unit 32 is stored in the trained model storage unit 34. When new information is input to the learning device 30 and new training data is acquired by the training data acquisition unit 31, the trained model Tm1 stored in the trained model storage unit 34 is updated appropriately according to the contents of the training data. In addition, the trained model Tm1 is transmitted from the learning device 30 to the anomaly detection device 40 described below, and is also stored in a trained model storage unit 45 of the anomaly detection device 40.
[0113] <How to generate a trained model> Next, a method for generating the trained model Tm1 by the learning device 30 will be described. The method for generating the trained model Tm1 includes a training data acquisition step and a learning calculation step. These steps are realized by a computer device constituting the learning device 30 executing a predetermined program.
[0114] First, when the training data acquisition process is started, the training data acquisition unit 31 acquires multiple sets of training data. For example, an operator artificially clogs the gate 245 and inputs the manner in which the gate 245 is clogged into the input unit 50, so that the training data acquisition unit 31 acquires gate information. Then, a molded product is molded using the gate 245, and the time point X1, the first evaluation value A1, and the second evaluation value A2 corresponding to the gate information are acquired based on the time series data of the pressure detected by the pressure sensor 25 at that time. This completes the training data acquisition process.
[0115] Next, when the learning calculation process is started, the learning calculation unit 32 learns the correspondence between the input information and the gate information based on multiple sets of training data, and generates a learned model Tm1. The learned model Tm1 is stored in the learned model storage units 34 and 45. This completes the learning calculation process.
[0116] <Explanation of the anomaly detection device> 10 is a block diagram showing the functional configuration of an abnormality detection device 40 according to this embodiment. The abnormality detection device 40 has a data acquisition unit 41, an abnormality detection unit 42, an output unit 43, a molding information storage unit 44, and a trained model storage unit 45. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as a HDD. The calculation unit executes data acquisition processing and abnormality detection processing, which will be described later, based on a program stored in the storage unit.
[0117] The molding information storage unit 44 stores molding information in a table format in which various first information and second information are associated with each other, similar to the molding information storage unit 33. The trained model storage unit 45 stores the trained model Tm1 generated by the learning device 30.
[0118] The forming information storage unit 44 and the trained model storage unit 45 may be realized in the same storage area as the forming information storage unit 33 and the trained model storage unit 34 of the learning device 30 in the computer device, or may be realized in a different storage area. That is, the learning device 30 and the anomaly detection device 40 may be configured to share the same forming information storage unit 33 and trained model storage unit 34, or the learning device 30 and the anomaly detection device 40 may be configured to have independent forming information storage units 33, 44 and trained model storage units 34, 45, respectively.
[0119] The data acquisition unit 41 executes a data acquisition process to acquire information for performing anomaly detection from each part of the anomaly detection system 10. The information for performing anomaly detection is, for example, a set of time point X1, a first evaluation value A1, a second evaluation value A2, an environmental value, and molding information acquired when molding the molded product to be detected (hereinafter, these pieces of information are collectively referred to as a "detection data set").
[0120] The abnormality detection unit 42 inputs the detection data set acquired by the data acquisition unit 41 to the trained model Tm1. Based on these inputs, the trained model Tm1 outputs detection information D1 for detecting a blockage of the gate 245.
[0121] The detection information D1 is information including the accuracy of each piece of gate information. Specifically, the detection information D1 includes a first accuracy which is the accuracy that the gate 245 is completely clogged, a second accuracy which is the accuracy that the gate 245 is partially clogged, a third accuracy which is the accuracy that the gate 245 is temporarily clogged, and a fourth accuracy which is the accuracy that the gate 245 is not clogged. As an example, when a detection dataset is input to the trained model Tm1, the trained model Tm1 outputs the detection information D1 with a first accuracy of 10%, a second accuracy of 10%, a third accuracy of 10%, and a fourth accuracy of 70%.
[0122] The output unit 43 determines the manner in which the gate 245 is clogged based on the detection information D1 acquired by the abnormality detection unit 42. The output unit 43 acquires, for example, the most probable manner of clogging from the detection information D1 as the determination result of the manner in which the gate 245 is clogged. For example, if the fourth probability is the highest from the above-mentioned first to fourth probabilities, the output unit acquires "no clogging has occurred in the gate 245" as the determination result.
[0123] The output unit 43 outputs the determination result to the display unit 60 and the control unit 271. The determination result is displayed on the display unit 60. If it is determined that the gate 245 is completely clogged, partially clogged, or temporarily clogged, the determination result may be displayed in a highlighted color such as red on the display of the display unit 60, and an alert may be issued from the speaker.
[0124] Furthermore, when it is determined that the gate 245 is completely or partially clogged, there is a high possibility that the gate 245 will remain clogged even after the molded product is molded. For this reason, when it is determined that the gate 245 is completely or partially clogged, the molding device 20 including the gate 245 determined to be clogged may be configured to stop with the mold section 24 open by an operation command from the control section 271. In this case, the operator inspects the mold section 24 based on an alert or the like from the display section 60, and cleans or replaces the mold section 24 as necessary.
[0125] In this embodiment, the output unit 43 may not be provided, and the detection information D1 obtained by the abnormality detection unit 42 may be displayed directly on the display unit 60. In this case, the operator may determine whether or not there is a blockage in the gate 245 based on the detection information D1 displayed on the display unit 60.
[0126] <Method of detecting abnormalities using an abnormality detection device> Next, a description will be given of an anomaly detection method performed by the anomaly detection device 40. The anomaly detection method includes a data acquisition step and an anomaly detection step. These steps are realized by a computer device constituting the anomaly detection device 40 executing a predetermined program.
[0127] When the data acquisition process is started, the data acquisition unit 41 acquires a detection data set that is acquired when the molded product to be detected is molded. This ends the data acquisition process. Next, when the anomaly detection process is started, the anomaly detection unit 42 acquires detection information D1 by inputting the detection data set to the trained model Tm1. Next, the output unit 43 acquires a judgment result based on the detection information D1. Finally, the output unit 43 outputs the judgment result to the display unit 60 and the control unit 271. This ends the anomaly detection process.
[0128] <Actions and Effects of the Anomaly Detection System> The abnormality detection system 10 of this embodiment acquires a first evaluation value A1 and a second evaluation value A2 based on time series data of pressure detected by the pressure sensor 25, and acquires detection information D1 for detecting blockage of the gate 245 based on the first evaluation value A1 and the second evaluation value A2 and the learned model Tm1.
[0129] With this configuration, once the learned model Tm1 is generated, it becomes possible to detect clogging of the gate 245 from information obtained by various sensors (pressure sensor 25, pressurization sensor 227, etc.) provided in the molding apparatus 20.
[0130] Furthermore, the abnormality detection system 10 according to this embodiment acquires detection information D1 by inputting the first evaluation value A1 and the second evaluation value A2 to a trained model Tm1 that has been machine-learned to learn the correlation between the first evaluation value A1 and the second evaluation value A2 and the gate information. Here, explanatory variables of the trained model Tm1 include the first evaluation value A1 and the second evaluation value A2, and a target variable of the trained model Tm1 includes the gate information. With this configuration, clogging of the gate 245 can be detected more accurately even when molding conditions vary.
[0131] In particular, conventional detection methods are unable to detect cases where temporary clogging occurs in the gate 245. In contrast, the abnormality detection system 10 according to this embodiment separates an area (from the first time point to the second time point) where the influence of the pressure rise due to temporary clogging is large, from an area (from the third time point to the fourth time point) where the influence is small, and detects clogging of the gate 245 based on a first evaluation value A1 and a second evaluation value A2 which are the time integrals of the pressure in each area. This makes it possible to detect temporary clogging of the gate 245 which could not be detected conventionally, and more reliably detect clogging of the gate 245.
[0132] <Second embodiment> The anomaly detection system according to the first embodiment has been described above. However, the implementation of the present invention is not limited to this, and various modifications can be made. Below, an anomaly detection system 11 according to a second embodiment of the present invention will be described. In the following description, the same reference numerals are used for parts that are unchanged from the first embodiment, and descriptions thereof will be omitted.
[0133] 11 is a block diagram illustrating a schematic diagram of an abnormality detection system 11 according to the second embodiment. The abnormality detection system 11 includes a plurality of molding devices 20, an abnormality detection device 40a, an input unit 50, and a display unit 60.
[0134] In this embodiment, the abnormality detection device 40a of the abnormality detection system 11 detects that there is a temporary blockage in the gate 245 when the ratio (A2 / A1) of the first evaluation value A1 to the second evaluation value A2 falls outside a predetermined range. That is, the abnormality detection system 11 differs from the abnormality detection system 10 according to the first embodiment in that it detects an abnormality by comparing the ratio (A2 / A1) with a predetermined range without using the trained model Tm1.
[0135] 12 is a block diagram showing the functional configuration of an abnormality detection device 40a according to this embodiment. The abnormality detection device 40a has a data acquisition unit 41, an abnormality detection unit 42a, an output unit 43a, a molding information storage unit 44, and a reference value storage unit 46. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as a HDD. The data acquisition unit 41 acquires a first evaluation value A1 and a second evaluation value A2 in the same manner as the data acquisition unit 41 according to the first embodiment.
[0136] A predetermined range is stored in the reference value storage unit 46. The predetermined range is a range defined based on the ratio between the first evaluation value A1 and the second evaluation value A2 acquired when there is no blockage in any of the multiple gates 245. The predetermined range is, for example, a range defined by an upper limit straight line Lmax and a lower limit straight line Lmin shown in FIG.
[0137] The abnormality detection unit 42a acquires detection information D2 for detecting clogging of the gate 245 by comparing the first evaluation value A1 and the second evaluation value A2 acquired by the data acquisition unit 41 with a predetermined range stored in the reference value storage unit 46. The detection information D2 is, for example, the difference (A2-Lmin(A1)) between the second evaluation value A2 and the point Lmin(A1) on the lower limit line Lmin corresponding to the first evaluation value A1.
[0138] The output unit 43a judges whether or not there is a temporary blockage in the gate 245 based on the detection information D2 acquired by the abnormality detection unit 42a. For example, when the detection information D2 is the above-mentioned difference (A2-Lmin(A1)), the output unit 43a judges that there is a temporary blockage in the gate 245 when the detection information D2 is a negative value (i.e., when the point on which the first evaluation value A1 and the second evaluation value A2 are plotted is located below the lower limit line Lmin). The output unit 43a outputs the judgment result to the display unit 60 and the control unit 271.
[0139] In this embodiment, the output unit 43a may not be provided, and the detection information D2 obtained by the abnormality detection unit 42a may be directly displayed on the display unit 60. In this case, the operator may determine whether the gate 245 is clogged based on the detection information D2 displayed on the display unit 60.
[0140] According to the abnormality detection system 11 of this embodiment, a temporary blockage of the gate 245 can be easily detected by comparing the first evaluation value A1 and the second evaluation value A2 with a predetermined range.
[0141] <Other> The embodiments disclosed above are illustrative in all respects and are not restrictive. In other words, the anomaly detection system of the present invention is not limited to the embodiments shown in the drawings, and may have other embodiments within the scope of the present invention. [Explanation of symbols]
[0142] 10 Abnormality detection system 11 Abnormality detection system 20 Molding device 21 bed 22 ejection section 221 hopper 222 Cylinder 223 Screw 224 Nozzle 225 Ball screw 226 Motor 227 Pressure sensor 228 Movement sensor 229 Heater 23 Mold clamping section 231 Fixed plate 231a Through hole 232 Movable plate 232a through hole 233 tie bar 234 ball screw 235 Support plate 236 Clamping force sensor 237 Motor 24 Mold section 241 Mold 242 Mold 243 Spool 243a End 243b End 244 Runner 245 Gate 25 Pressure sensor 26 Temperature sensor 27 Control panel 271 Control unit 272 communication unit 30 learning device 31 training data acquisition unit 32 Learning calculation unit 33 Molding information storage unit 34 Trained model storage unit 40 Anomaly detection device 40a Anomaly detection device 41 data acquisition unit 42 abnormality detection unit 42a abnormality detection unit 43 output unit 43a output unit 44 molding information storage unit 45 Trained model memory unit 46 Reference value memory unit 50 Input unit 60 Display C1 Cavity GT1~GT4 Gate Time X1 (first time point) Time X2 (second time point) Time X3 (third time point) X4 time point (fourth time point) Fm1 Foreign object A1 First evaluation value A2 Second evaluation value D1 Detection information D2 Detection information
Claims
1. An abnormality detection system including a molding device that molds a molded product and an abnormality detection device that detects an abnormality in the molding device, The molding device includes: a mold part having a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates formed therein; an injection unit that performs a filling operation of filling the cavity with a molten molding material from the spool, a pressure holding operation of holding the pressure of the molding material filled in the cavity, and a pressure holding release operation of releasing the pressure holding of the molding material; a pressure sensor for detecting a pressure of the molding material at the spool, the runner, or the gate; having The abnormality detection device includes: a data acquisition unit that acquires a first evaluation value obtained by integrating, with time, the pressure detected by the pressure sensor from a first time point after the filling operation to a second time point before the pressure-holding release operation, and a second evaluation value obtained by integrating, with time, the pressure detected by the pressure sensor from a third time point after the first time point to a fourth time point after the second time point and before the pressure-holding release operation; an abnormality detection unit that acquires detection information for detecting a blockage of the gate based on the first evaluation value and the second evaluation value; having The first time point is a time point at which the pressure holding operation is performed or a time point at which a rise in the pressure is detected, The fourth time point is a time point of the pressure release operation, The second time point is a time point before an intermediate point between the first time point and the fourth time point, The anomaly detection system, wherein the third point in time is a point in time after the intermediate point.
2. the abnormality detection unit acquires the detection information based on a ratio between the first evaluation value and the second evaluation value. The anomaly detection system of claim 1 .
3. the abnormality detection unit acquires the detection information indicating that the gate is clogged when the ratio acquired during molding of the molded product to be detected falls outside a predetermined range defined based on the ratio acquired when none of the plurality of gates is clogged. The anomaly detection system of claim 2 .
4. The abnormality detection unit acquires the detection information by inputting the first evaluation value and the second evaluation value to a trained model that has been machine-learned to learn a correlation between the first evaluation value, the second evaluation value, and the clogging of the gate. The anomaly detection system of claim 1 .
5. The pressure sensor is provided in a cold slug well. The anomaly detection system according to any one of claims 1 to 4.
6. An abnormality detection device for detecting an abnormality in a molding device that molds a molded product, The molding device includes: a mold part having a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates formed therein; an injection unit that performs a filling operation of filling the cavity with a molten molding material from the spool, a pressure holding operation of holding the pressure of the molding material filled in the cavity, and a pressure holding release operation of releasing the pressure holding of the molding material; a pressure sensor for detecting a pressure of the molding material at the spool, the runner, or the gate; having The abnormality detection device includes: a data acquisition unit that acquires a first evaluation value obtained by integrating, with time, the pressure detected by the pressure sensor from a first time point after the filling operation to a second time point before the pressure-holding release operation, and a second evaluation value obtained by integrating, with time, the pressure detected by the pressure sensor from a third time point after the first time point to a fourth time point after the second time point and before the pressure-holding release operation; an abnormality detection unit that acquires detection information for detecting a blockage of the gate based on the first evaluation value and the second evaluation value; Equipped with The first time point is a time point at which the pressure holding operation is performed or a time point at which a rise in the pressure is detected, The fourth time point is a time point of the pressure release operation, The second time point is a time point before an intermediate point between the first time point and the fourth time point, The third point in time is a point in time after the intermediate point in time.
7. 1. An abnormality detection method for detecting an abnormality in a molding device including a mold section having a spool, a plurality of runners branching off from the spool, a plurality of gates respectively connected to the plurality of runners, and a cavity connected to the plurality of gates, and an injection section which performs a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, a data acquiring process for acquiring a first evaluation value obtained by integrating, with time, a pressure detected by a pressure sensor that detects the pressure of the molding material at the spool, the runner, or the gate from a first point in time after the filling operation to a second point in time before the holding pressure release operation, and a second evaluation value obtained by integrating, with time, a pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the holding pressure release operation; an abnormality detection process of acquiring detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value; Equipped with The first time point is a time point at which the pressure holding operation is performed or a time point at which a rise in the pressure is detected, The fourth time point is a time point of the pressure release operation, The second time point is a time point before an intermediate point between the first time point and the fourth time point, The anomaly detection method, wherein the third time point is a time point after the intermediate point.
8. 1. A program for detecting an abnormality in a molding device comprising: a mold section having a spool, a plurality of runners branching off from the spool, a plurality of gates each connected to the runners, and a cavity connected to the plurality of gates formed therein; and an injection section which performs a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, a data acquiring process for acquiring a first evaluation value obtained by integrating, with time, a pressure detected by a pressure sensor that detects the pressure of the molding material at the spool, the runner, or the gate from a first point in time after the filling operation to a second point in time before the holding pressure release operation, and a second evaluation value obtained by integrating, with time, a pressure detected by the pressure sensor from a third point in time after the first point in time to a fourth point in time after the second point in time and before the holding pressure release operation; an abnormality detection process of acquiring detection information for detecting clogging of the gate based on the first evaluation value and the second evaluation value; on a computer device, The first time point is a time point at which the pressure holding operation is performed or a time point at which a rise in the pressure is detected, The fourth time point is a time point of the pressure release operation, The second time point is a time point before an intermediate point between the first time point and the fourth time point, The program, wherein the third point in time is a point in time after the intermediate point.
9. A trained model for detecting an abnormality in a molding device including: a mold section having a spool, a plurality of runners branching off from the spool, a plurality of gates each connected to the plurality of runners, and a cavity connected to the plurality of gates formed therein; and an injection section which performs a filling operation for filling a molten molding material from the spool into the cavity, a pressure holding operation for holding the pressure of the molding material filled in the cavity, and a pressure holding release operation for releasing the pressure holding of the molding material, The explanatory variables are: a first evaluation value obtained by time-integrating a pressure detected by a pressure sensor that detects a pressure of the molding material in the spool, the runner, or the gate during a period from a first time point after the filling operation to a second time point before the pressure-holding release operation; and a second evaluation value obtained by integrating the pressure detected by the pressure sensor over time from a third time point that is later than the first time point to a fourth time point that is later than the second time point and before the pressure-holding release operation, The objective variable contains the value of the gate blockage, The first time point is a time point at which the pressure holding operation is performed or a time point at which a rise in the pressure is detected, The fourth time point is a time point of the pressure release operation, The second time point is a time point before an intermediate point between the first time point and the fourth time point, The third time point is a time point after the intermediate point. Trained model.
Citation Information
Patent Citations
Monitoring method for injection molding machine
JP1991211032A
Device and method for controlling injection molding
JP2000052396A
Injection molding machine and quality judging method of molded product
JP2005161557A
Injection controlling method of injection molding machine
JP2006341540A
Injection molding machine and method for detecting abnormality of same
JP2009000929A
