WO2021206017A1 - Molding system, irregularity prediction device, irregularity prediction method, program, and trained model - Google Patents

Molding system, irregularity prediction device, irregularity prediction method, program, and trained model Download PDF

Info

Publication number
WO2021206017A1
WO2021206017A1 PCT/JP2021/014318 JP2021014318W WO2021206017A1 WO 2021206017 A1 WO2021206017 A1 WO 2021206017A1 JP 2021014318 W JP2021014318 W JP 2021014318W WO 2021206017 A1 WO2021206017 A1 WO 2021206017A1
Authority
WO
WIPO (PCT)
Prior art keywords
pressure
molding
molding material
evaluation value
molded product
Prior art date
Application number
PCT/JP2021/014318
Other languages
French (fr)
Japanese (ja)
Inventor
智也 足立
紀行 馬場
慎太郎 辻
勇佐 大久保
幸治 木村
Original Assignee
株式会社ジェイテクト
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社ジェイテクト filed Critical 株式会社ジェイテクト
Publication of WO2021206017A1 publication Critical patent/WO2021206017A1/en

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
    • B29C45/17—Component parts, details or accessories; Auxiliary operations
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C45/00—Injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould; Apparatus therefor
    • B29C45/17—Component parts, details or accessories; Auxiliary operations
    • B29C45/76—Measuring, controlling or regulating

Definitions

  • This disclosure relates to molding systems, anomaly predictors, anomaly prediction methods, programs and trained models.
  • An injection molding apparatus in which a melt obtained by melting a molding material is supplied to a cavity formed between a plurality of molds to mold a molded product.
  • Examples of defective molded products include burrs. The occurrence of burrs not only spoils the appearance of the molded product, but may also affect the performance of the molded product. For this reason, we are currently visually inspecting molded products.
  • Patent Document 1 in order for the operator to judge the suitability of the current injection conditions with respect to the occurrence of burrs, a curve showing the time-dependent change of the pressure in the mold detected by the pressure sensor and the pressure as a guide for the burrs to blow are described.
  • a technique for drawing a burr blowing limit curve on a display unit is disclosed.
  • the burr blowing limit curve is a curve that can be drawn by acquiring various values by experiments.
  • the burr blowing limit curve according to Patent Document 1 is a different curve depending on the molding conditions. Molding conditions include the shape of the mold, the type / viscosity / temperature of the molding material, the pressure applied to the molding material (melt), the molding environment (room temperature, humidity), and the like. Therefore, in order to determine the occurrence of burrs based on the technique according to Patent Document 1, it is necessary to prepare a burr blowing limit curve for each molding condition. In addition, if the actual molding conditions deviate from the conditions that form the basis of the burr blowing limit curve, the actual pressure will be lower than the burr blowing limit curve even though burrs have occurred, and the occurrence of burrs will be accurate. It may not be possible to judge.
  • an object of the present disclosure is to provide a molding system, an abnormality prediction device, an abnormality prediction method, a program, and a trained model for more accurately predicting the state of burrs.
  • the present disclosure is a molding system including a molding device and an abnormality predicting device for predicting an abnormality of a molded product molded by the molding device, and the molding device combines a plurality of molds.
  • a mold clamping portion for tightening the plurality of molds, a filling operation for filling a cavity formed between the plurality of molds by tightening the plurality of molds with a molten molding material, and the above-mentioned
  • the injection portion that performs the pressure holding operation for holding the pressure of the molding material filled in the cavity and the pressure holding release operation for releasing the pressure holding of the molding material, and the pressure of the molding material in the cavity.
  • It has a pressure sensor to detect, and the abnormality predictor determines an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation based on the time series data of the pressure detected by the pressure sensor.
  • a molding system having a data acquisition unit to be acquired and an abnormality prediction unit for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value.
  • the inventors monotonically decreased the pressure in the cavity after the pressure release operation during normal times (when burrs did not occur), while the pressure increased or decreased momentarily when burrs occurred.
  • the degree of decrease becomes smaller (the pressure gradient becomes larger for a moment). That is, it was discovered that the state of burrs can be predicted by paying attention to the change in the pressure of the molding material after the pressure holding release operation. Therefore, the molding system according to the present disclosure acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
  • the present disclosure is the molding system described in (1) above, and the abnormality prediction unit uses a trained model in which the correlation between the evaluation value and the burr state of the molded product is machine-learned.
  • the prediction information is acquired, the explanatory variable of the trained model includes the evaluation value, and the objective variable of the trained model is the burr state of the molded product.
  • the burr state may provide a molding system including the presence or absence of the burr or the size of the burr.
  • the burr state can be predicted more accurately even if the molding conditions vary.
  • the present disclosure is the molding system according to (2) above, wherein the molding apparatus further includes a temperature sensor for detecting the temperature of the molding material, and the explanatory variables of the trained model are: Molding that includes a second evaluation value related to the temperature detected by the temperature sensor, and the abnormality prediction unit acquires the prediction information by inputting the evaluation value and the second evaluation value into the trained model.
  • the molding apparatus further includes a temperature sensor for detecting the temperature of the molding material
  • the explanatory variables of the trained model are: Molding that includes a second evaluation value related to the temperature detected by the temperature sensor, and the abnormality prediction unit acquires the prediction information by inputting the evaluation value and the second evaluation value into the trained model.
  • a system may be provided.
  • the change in the pressure of the molding material after the pressure holding release operation also depends on the temperature of the molding material.
  • the higher the temperature of the molding material the lower the viscosity of the molding material and the easier it is for the molding material to flow. Therefore, the higher the temperature of the molding material, the more likely it is that the pressure change at the time of burr generation will occur sharply within a short period of time.
  • the present disclosure is the molding system according to (1) above, and the abnormality prediction unit is more than a reference value generated based on the evaluation value acquired at the time of normal molding of the molded product.
  • the present disclosure is the molding system according to any one of (1) to (4) above, and the data acquisition unit is a pressure obtained based on the time series data after the pressure holding release operation.
  • the evaluation value is acquired based on the second time-series data of the inclination of, and the evaluation value is the maximum value of the inclination in the second time-series data or the inclination of 0 in the second time-series data.
  • the present disclosure is the molding system according to any one of (1) to (4) above, and the data acquisition unit maximizes the pressure at the minimum of the time series data after the pressure holding release operation.
  • a molding system may be provided that acquires, as the evaluation value, the difference from the pressure at the above, or the rate of change from the pressure at the minimum to the pressure at the maximum. With this configuration, the evaluation value can be easily obtained from the time series data.
  • the present disclosure is an abnormality prediction device for predicting an abnormality in a molded product molded by a molding device, wherein the molding device is a mold for tightening the plurality of molds in a state where a plurality of molds are combined.
  • the filling operation of filling the molded material in a molten state into the tightening portion and the cavity formed between the plurality of molds by tightening the plurality of molds, and the pressure of the molding material filled in the cavity. It has an injection unit that performs a holding pressure holding operation and a pressure holding release operation that releases the pressure holding of the molding material, and a pressure sensor that detects the pressure of the molding material in the cavity.
  • the abnormality prediction device includes a data acquisition unit that acquires an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation based on the time-series data of the pressure detected by the pressure sensor, and the evaluation value. Based on this, the present invention provides an abnormality prediction device including an abnormality prediction unit for acquiring prediction information for predicting a burr state of the molded product.
  • the abnormality prediction device acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
  • a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material.
  • It is an abnormality prediction method for predicting an abnormality of a molded product to be formed, and is after the pressure holding release step based on time-series data of pressure detected by a pressure sensor that detects the pressure of the molding material in the cavity. It includes a data acquisition step of acquiring an evaluation value regarding a change in pressure of the molding material, and an abnormality prediction step of acquiring prediction information for predicting a burr state of the molded product based on the evaluation value.
  • An abnormality prediction method is provided.
  • the abnormality prediction method acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
  • a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material. It is a program for predicting an abnormality of a molded product to be formed, and is after the pressure holding release step based on time-series data of pressure detected by a pressure sensor that detects the pressure of the molding material in the cavity.
  • a computer device includes a data acquisition process for acquiring an evaluation value regarding a change in pressure of the molding material, and an abnormality prediction process for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value. Provide a program to be executed by.
  • a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step In a state where a plurality of molds are combined, a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step.
  • Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material.
  • It is a trained model for predicting the abnormality of the molded product to be made, and the explanatory variables are acquired based on the time-series data of the pressure detected by the pressure sensor that detects the pressure of the molding material in the cavity.
  • the objective function is the burr state of the molded product, and the burr state is the presence or absence of the burr or the burr.
  • the explanatory function is the evaluation value regarding the change in the pressure of the molding material after the pressure release operation and the objective function is the burr state of the molded product
  • the evaluation value is input to the trained model.
  • the predicted burr state is output.
  • the state of burrs can be predicted based on the change in the pressure of the molding material after the pressure holding release operation.
  • the state of burrs can be predicted more accurately.
  • FIG. 1 is a block diagram schematically showing a molding system according to the first embodiment.
  • FIG. 2 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1.
  • FIG. 3 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1.
  • FIG. 4 is a cross-sectional view of a mold portion schematically showing a state at the end of the filling process.
  • 5 (a) and 5 (b) are examples of graphs showing time-series data of pressure and pressure gradient.
  • FIG. 6 is a cross-sectional view of a mold portion schematically showing a state at the end of a filling process when burrs are generated in a molded product.
  • FIG. 1 is a block diagram schematically showing a molding system according to the first embodiment.
  • FIG. 2 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1.
  • FIG. 3 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1.
  • FIG. 4 is a cross-sectional view of
  • FIG. 7 is a cross-sectional view schematically showing a state of the pressure holding release step when burrs are generated in the molded product.
  • 8 (a) and 8 (b) are examples of graphs showing time-series data of pressure and pressure gradient when burrs are generated in the molded product.
  • FIG. 9 is a block diagram showing a functional configuration of the learning device according to the first embodiment.
  • 10 (a) and 10 (b) are explanatory views for explaining the evaluation values according to the first embodiment.
  • FIG. 11 is a block diagram showing a functional configuration of the abnormality prediction device according to the first embodiment.
  • FIG. 12 is a block diagram schematically showing the molding system according to the second embodiment.
  • FIG. 13 is a block diagram showing a functional configuration of the abnormality prediction device according to the second embodiment.
  • FIG. 14 is a graph schematically explaining the comparison between the evaluation value and the attention reference value according to the second embodiment.
  • FIG. 1 is a block diagram schematically showing a molding system 10 according to the first embodiment.
  • the molding system 10 includes a plurality of molding devices 20, a learning device 30, an abnormality prediction device 40, an input unit 50, and a display unit 60.
  • the molding device 20, the learning device 30, the abnormality prediction device 40, the input unit 50, and the display unit 60 are each provided so as to be able to communicate wirelessly or by wire.
  • the learning device 30 and the abnormality prediction device 40 are composed of an information processing device (computer device) having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.).
  • the learning device 30 and the abnormality prediction device 40 may be configured by the same information processing device or may be configured by separate information processing devices.
  • a plurality of molding devices 20 are connected to one learning device 30 and one abnormality prediction device 40, and the learning device 30 and the abnormality prediction device 40 are used for various data transmitted from the plurality of molding devices 20. Based on this, learning and abnormality prediction are performed.
  • the molding device 20 may have a one-to-one correspondence with the learning device 30 and the abnormality prediction device 40. That is, in the molding system 10, the molding device 20 may be one, or a plurality of learning devices 30 and abnormality prediction devices 40 may be provided.
  • the input unit 50 is, for example, a keyboard or a mouse, and receives various inputs from the operator.
  • the display unit 60 is, for example, a display or a speaker, and displays various information in the molding system 10.
  • the input unit 50 and the display unit 60 may be integrated, for example, a touch panel. Further, the input unit 50 and the display unit 60 may be provided as a portable terminal device so as to be movable to a place away from the molding device 20, the learning device 30, and the abnormality prediction device 40.
  • FIGS. 2 and 3 are explanatory views conceptually showing the molding apparatus 20.
  • the molding apparatus 20 includes a bed 21, an injection portion 22, a mold clamping portion 23, a mold portion 24, a pressure sensor 25, a temperature sensor 26, and a control panel 27.
  • FIG. 2 shows a molding apparatus 20 in a state where the mold portion 24 is open
  • FIG. 3 shows a molding apparatus 20 in a state where the mold portion 24 is combined.
  • the molding apparatus 20 is an apparatus for performing mold-fastening injection molding.
  • 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 apparatus 20 and outputs an operation command to each drive unit. Further, the control unit 271 is electrically connected to each sensor (pressure sensor 25 or the like) of the molding apparatus 20, and the signal detected by each sensor is input to the control unit 271.
  • the control unit 271 is composed of an information processing device having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.).
  • the communication unit 272 communicates with each other unit (learning device 30, etc.) of the molding system 10.
  • the communication unit 272 transmits, for example, the signal detected by each of the sensors to the learning device 30 or the abnormality prediction device 40. Further, the communication unit 272 receives the determination information and the prediction information described later from the abnormality prediction device 40.
  • the mold clamping portion 23 includes a fixed plate 231, a movable plate 232, a tie bar 233, a ball screw 234, a support plate 235, a mold clamping force sensor 236, and a motor 237.
  • the fixing plate 231 and the supporting plate 235 are fixed to the bed 21.
  • the support board 235 supports the ball screw 234.
  • the ball screw 234 is connected to the motor 237. When the motor 237 is rotated by the operation command of the control unit 271, the ball screw 234 moves.
  • a movable platen 232 is fixed to the end of the ball screw 234 opposite to the end connected to the motor 237.
  • the direction in which the ball screw 234 moves is referred to as an "axial direction".
  • the side where the motor 237 is located with respect to the ball screw 234 is referred to as the "one side” in the axial direction
  • the side where the movable platen 232 is located with respect to the ball screw 234 is referred to as the “other side” in the axial direction.
  • the movable board 232 moves in the axial direction as the ball screw 234 moves.
  • the movable platen 232 is formed with a through hole 232a penetrating in the axial direction.
  • the end of the tie bar 233 on one side in the axial direction is fixed to the support plate 235, and the end on the other side in the axial direction is fixed to the fixing plate 231.
  • the tie bar 233 is inserted into the through hole 232a of the movable platen 232. As a result, the tie bar 233 guides the axial movement of the movable platen 232.
  • the mold clamping force sensor 236 detects the pressure (reaction force of the mold clamping force) applied to the support plate 235 from the ball screw 234.
  • the mold clamping force sensor 236 outputs a detection signal regarding pressure to the control unit 271.
  • the mold clamping force sensor 236 may be installed at another position as long as it can detect the mold clamping force in the mold portion 24 described later.
  • the fixing plate 231 is formed with a through hole 231a having a diameter widened on the other side in the axial direction. A cylinder 222, which will be described later, is inserted into the through hole 231a.
  • the mold unit 24 has a plurality of molds 241 and 242.
  • the mold 241 is fixed to the movable platen 232.
  • the mold 241 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 fixing plate 231. That is, the mold 242 is a fixed mold.
  • a flow path 243 is formed in the mold 242.
  • the injection unit 22 includes a hopper 221, a cylinder 222, a screw 223, a ball screw 225, a motor 226, a pressurization sensor 227, and a heater 228.
  • the hopper 221 is connected to the cylinder 222 and supplies the 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 one end of the cylinder 222 on one side in the axial direction becomes narrower as it approaches the end on one side in the radial direction, and a nozzle 224 is provided at the end.
  • the nozzle 224 is connected to the flow path 243 of the mold 242.
  • the screw 223 is inserted into the cylinder 222 from the end on the other side in the axial direction of the cylinder 222.
  • a ball screw 225 is connected to the other side of the screw 223 in the axial direction, and a motor 226 is connected to the other side of the ball screw 225 in the axial direction.
  • the motor 226 is rotated by the operation command of the control unit 271
  • the ball screw 225 moves in the axial direction.
  • the screw 223 also moves in the axial direction.
  • the screw 223 rotates in the circumferential direction with the axial direction as the central axis.
  • the pressurization sensor 227 detects the pressure applied from the ball screw 225 to the motor 226 (the reaction force of the pushing force of the screw 223).
  • the pressurization sensor 227 outputs a detection signal related to pressure to the control unit 271.
  • the pressurization sensor 227 may be installed at another position as long as it can detect the pushing force of the screw 223.
  • the heater 228 is, for example, a resistance heating heater in which a resistance wire is wound in a coil shape.
  • the heater 228 heats the inside of the cylinder 222 by the resistance heat when a current is passed through the resistance wire by the operation command of the control unit 271.
  • the pressure sensor 25 is installed in the region of the molds 241 and 242 facing the cavity C1.
  • the pressure sensor 25 detects the pressure in the cavity C1.
  • the pressure sensor 25 detects the pressure of the molding material (melted state or solidified state, or a state in which the molten state and the solidified state are mixed) supplied into the cavity C1.
  • the pressure sensor 25 outputs a detection signal related to pressure to the control unit 271.
  • a plurality of pressure sensors 25 are installed on both the molds 241 and 242, respectively. However, only one pressure sensor 25 may be installed in one of the molds 241 and 242.
  • the temperature sensor 26 is built in the mold 241 and detects the temperature of the mold 241.
  • the temperature sensor 26 outputs a detection signal related to temperature to the control unit 271.
  • the temperature sensor 26 may be installed in a region of the mold 241 facing the cavity C1 or may be installed in the mold 242. Further, the temperature sensor 26 may be installed in the cylinder 222. That is, the temperature sensor 26 may be able to directly or indirectly detect the temperature of the molding material supplied into the cavity C1.
  • a method for manufacturing a molded product by the molding apparatus 20 will be described with reference to FIGS. 2 to 5 (b) as appropriate.
  • the pre-process ST1, the mold clamping process ST2, the filling process ST3, the pressure holding process ST4, the pressure holding release process ST5, and the mold release process ST6 are performed in this order. Will be executed.
  • the molded product is a resin cage used for rolling bearings.
  • this is an example of a molded product, and the molded product molded by the molding apparatus according to the present disclosure may be a molded product having another shape and use.
  • the pre-process ST1 is executed.
  • the screw 223 is rotated by the motor 226, and the 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 228.
  • the pellets of the molding material are melted in the cylinder 222 by the frictional heat accompanying the rotation of the screw 223 and the heating by the heater 228 to become the molding material L1 in the molten state.
  • the previous step ST1 is completed.
  • the ball screw 234 is moved to the other side in the axial direction by the operation command of the control unit 271, and the mold is as shown in FIG.
  • the 241 is brought into contact with the mold 242.
  • the ball screw 234 further presses the mold 241 against the mold 242 toward the other side in the axial direction by a predetermined mold tightening force. That is, the plurality of molds 241 and 242 are tightened.
  • the cavity C1 is formed between the plurality of molds 241 and 242.
  • the mold clamping force is one of the molding conditions, and is determined according to other molding conditions such as the shape of the molds 241 and 242.
  • the mold clamping force is detected by the mold clamping force sensor 236.
  • the ball screw 225 moves to one side in the axial direction while maintaining the above-mentioned mold clamping force.
  • the screw 223 pushes the molding material L1 to one side in the axial direction, and the molding material L1 is ejected from the nozzle 224 of the cylinder 222 to the cavity C1 via the flow path 243 of the mold 242 (filling operation).
  • FIG. 4 is a cross-sectional view of the mold portion 24 schematically showing the state at the end of the filling step ST3.
  • the flow path 243 has a first opening 243a that opens on the nozzle 224 side and a second opening 243b that opens on the cavity C1 side.
  • the cavity C1 is formed in an annular shape.
  • the molding material L1 more specifically, the molding material in a molten state
  • the filling step is performed.
  • ST3 ends.
  • the molten molding material L1 is supplied into the cavity C1 while gradually solidifying from the vicinity of the surface of the molds 241 and 242.
  • the screw 223 further pushes the molding material L1 to one side in the axial direction, and the molding material L1 is further injected from the nozzle 224 of the cylinder 222 into the cavity C1.
  • a predetermined pressure Pt1 for example, several tens to several hundreds of MPa
  • the screw 223 continues to apply a predetermined pressure to the molding material L1 for a predetermined time (for example, several seconds) (pressure holding operation).
  • the pressure (pressurization) at which the screw 223 pushes the molding material L1 into the cavity C1 is detected by the pressurization sensor 227.
  • the screw 223 moves to the other side in the axial direction to release the pressure holding of the molding material L1 (pressure holding release operation).
  • the holding pressure releasing step ST5 ends.
  • the mold release step ST6 the mold portion 24 is cooled, so that the molding material L1 in the cavity C1 is completely solidified, and a molded product is formed.
  • the ball screw 234 moves to one side in the axial direction, and the mold 241 separates from the mold 242, so that the molded product is taken out.
  • the cooling of the mold portion 24 may be started at the same time as the pressure holding release step ST5.
  • 5 (a) and 5 (b) show the time-series data of the pressure detected by the pressurization sensor 227 and the pressure sensor 25 in the filling step ST3, the pressure holding step ST4, and the pressure holding release step ST5, and the pressure.
  • This is an example of a graph showing time-series data of inclination.
  • 5 (a) and 5 (b) show graphs obtained when the molded product is molded without the occurrence of burrs.
  • the vertical axis is the pressure P and the horizontal axis is the time t.
  • the graph line F1 shown by the broken line is the time series data of the pressure detected by the pressurization sensor 227, and represents the pressure at which the screw 223 pushes the molding material L1 into the cavity C1.
  • the graph line F2 shown by the solid line is the time series data of the pressure detected by the pressure sensor 25, and represents the pressure of the molding material L1 (the state including at least one of the molten state and the solidified state) in the cavity C1. ..
  • the pressurization rises from 0 to the pressure Pt1 in the filling step ST3, is held at the pressure Pt1 for a predetermined time in the pressure holding step ST4, and is held in the pressure Pt1 in the holding pressure releasing step ST5.
  • the pressure drops from Pt1.
  • the holding pressure release step ST5 is started from the time point X1. That is, the holding pressure release operation is executed at the time point X1.
  • the pressure of the molding material L1 in the cavity C1 decreases monotonically after the time point X1.
  • FIG. 5B shows the graph line F2a of the slope of the pressure obtained by time-differentiating the graph line F2 after the time point X1. That is, the graph line F2a is a graph showing the change in the pressure of the molding material L1 in the cavity C1. After the time point X1, the graph line F2 decreases monotonically, so that the graph line F2a becomes a value smaller than 0.
  • FIG. 6 is a cross-sectional view of the mold portion 24 schematically showing the state at the end of the filling step ST3 when burrs are generated in the molded product.
  • the foreign matter Fm1 is sandwiched between the molds 241 and the molds 242, as shown in an enlarged manner at the bottom. Therefore, an unintended gap C2 is formed between the mold 241 and the mold 242 in addition to the cavity C1. Further, the volume of the cavity C1 is also increased by the width W1 at which the gap C2 is formed as compared with the example of FIG. Then, in the filling step ST3, the molding material L1 is filled in the cavity C1 and the gap C2.
  • FIG. 7 is a cross-sectional view schematically showing the state of the pressure holding release step ST5 when burrs are generated in the molded product.
  • FIG. 7 shows the same area as the enlarged view of FIG.
  • the pressure holding step ST4 when the predetermined pressure Pt1 is held by the molding material L1, the molding material L1 is exposed to the surfaces (for example, the mold 241) exposed to the cavities C1 and the gaps C2 of the molds 241 and 242.
  • the surface 241a is pressed.
  • the force with which the molding material L1 pushes the exposed surface 241a is larger by the amount of the gap C2 than when burrs are not generated.
  • the pressure holding release step ST5 When the holding of the pressure Pt1 of the molding material L1 is released by the pressure holding release step ST5, the force with which the molding material L1 pushes the exposed surface 241a or the like weakens, and the mold clamping force of the mold clamping portion 23 becomes larger than the pressing force. By becoming stronger, the exposed surface 241a moves in the direction indicated by the arrow AR1 in FIG. In FIG. 7, the position of the exposed surface 241a during the pressure holding step ST4 is indicated by a two-dot chain line, and the position of the exposed surface 241a after the pressure holding release step ST5 is indicated by a solid line. That is, the pressure holding release operation slightly tightens the mold 241 and reduces the volume of the cavity C1.
  • the volume of the gap C2 is also reduced.
  • the density of the molding material L1 in the region C1a near the exposed surface 241a becomes higher than that before the pressure holding release operation.
  • the pressure of the molding material L1 detected by the pressure sensor 25 increases as the density of the molding material L1 in the region C1a increases.
  • the graph line F3 shown by the solid line is the time series data of the pressure detected by the pressure sensor 25, and represents the pressure of the molding material L1 in the cavity C1.
  • the pressure of the molding material L1 detected by the pressure sensor 25 rises only momentarily after the pressure holding release operation (after the time point X1). That is, when burrs occur, the graph line F3 after the holding pressure release operation shows a tendency of non-monotonic decrease.
  • FIG. 8B shows the graph line F3a of the slope of the pressure obtained by time-differentiating the graph line F3 after the time point X1. That is, the graph line F3a is a graph showing the change in the pressure of the molding material L1 in the cavity C1. Since the graph line F3 increases for a moment and then decreases monotonically after the time point X1, the graph line F3a has a region having a value larger than 0.
  • the degree of decrease in the pressure of the molding material L1 in the cavity C1 after the pressure holding release operation is the degree of increase in the density of the molding material L1 in the cavity C1. If it is larger than, the pressure increase shown by the arrow AR2 in FIG. 8A may not occur.
  • the time-series data of the pressure when burrs occur after the pressure holding release operation includes an increase in the density of the molding material L1 in the region C1a, and FIG. Compared with the time-series data of the pressure shown in, there is a part where the pressure decrease becomes slow for a moment. That is, after the pressure holding release operation, the slope of the pressure when burrs are generated becomes a larger value than the slope of the pressure when burrs are not generated.
  • the inventors have found that the pressure of the molding material L1 in the cavity C1 after the pressure holding release operation decreases monotonically under normal conditions (when burrs do not occur), while it becomes monotonous. It was discovered that when burrs occur, the pressure of the molding material L1 increases, does not decrease, or decreases for a moment (the pressure gradient increases for a moment). That is, it was discovered that the state of burrs can be predicted by paying attention to the change in the pressure of the molding material L1 after the pressure holding release operation.
  • the learned model Tm1 is trained in the learning device 30 to learn the correlation between the pressure change of the molding material L1 in the cavity C1 after the pressure holding release operation and the burr state.
  • the generation and the abnormality prediction device 40 acquire the prediction information for predicting the state of burrs based on the trained model Tm1 and the change in the pressure of the molding material L1 after the pressure holding release operation.
  • the learning device 30 and the abnormality prediction device 40 will be described.
  • FIG. 9 is a block diagram showing a functional configuration of the learning device 30 according to the present embodiment.
  • the learning device 30 includes 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 an HDD.
  • the molding information is, for example, table-type information in which various types of first information and second information are associated with each other.
  • first information is the type of the mold
  • second information includes various dimensions of the mold and the volume of the cavity C1.
  • the first information is the type or lot number of the molding material
  • the second information includes the physical properties (viscosity, moisture content, etc.) of the molding material.
  • the training data acquisition unit 31 acquires information on the training data from each unit of the molding system 10.
  • the training data includes an evaluation value described later, a second evaluation value, a third evaluation value, molding information, and burr information.
  • the training data also includes the pressures (molding force and pressurization) detected by the mold clamping force sensor 236 and the pressurization sensor 227, respectively.
  • the training data acquisition unit 31 acquires an evaluation value regarding a change in pressure after the pressurization release operation based on the pressure detected by the pressure sensor 25 and the pressurization sensor 227, respectively. In addition, the training data acquisition unit 31 acquires a second evaluation value regarding the temperature detected by the temperature sensor 26. In addition, the training data acquisition unit 31 acquires a third evaluation value regarding the surrounding and internal environment of the molding apparatus 20 detected by another sensor (for example, a humidity sensor) (not shown).
  • another sensor for example, a humidity sensor
  • FIG. 10A is a graph showing time-series data of the slope of pressure.
  • the vertical axis of FIG. 10A is the slope of pressure (dP / dt), and the horizontal axis is time t.
  • a plurality of pressure time series data are acquired.
  • the time-series data of the plurality of pressures acquired by the plurality of pressure sensors 25 are acquired, and a plurality of evaluation values are acquired respectively.
  • the time series data of the pressure obtained by one pressure sensor 25 among the plurality of pressure sensors 25 will be focused on.
  • the time series data of the inclination of the average pressure and one evaluation value are acquired. May be good.
  • FIG. 10A shows an example of time-series data of the slope of the pressure acquired when three molded products are molded.
  • the graph line F31 solid line
  • the graph line F32 two-dot chain line
  • the graph line F33 broken line
  • It is a graph which enlarges and shows the region where the slope becomes 0 or more in the slope of pressure (after X1).
  • the pressure gradient (dP / dt) at the time point t is, for example, the amount of change in pressure dP from the time point t to the time point (t + dt) in the time series data of pressure divided by the amount of change in time dt. Demanded by.
  • the evaluation value is, for example, the maximum value of the pressure gradient.
  • the evaluation value is dP1.
  • the evaluation value becomes dP2 or dP3.
  • the evaluation value is acquired by the training data acquisition unit 31 by calculating based on the pressure time series data.
  • the evaluation value is not limited to the slope of pressure (dP / dt), and may be the amount of change in pressure dP. Further, as shown in FIG. 10A, the evaluation value may be the full width at half maximum W1, W2, W3 of the graph lines F31, F32, and F33, respectively.
  • the full width at half maximum W1 is the width of the graph line F31 at the time when the slope of the pressure becomes half of the maximum value dP1 (dP1 / 2). That is, the evaluation value may be the width of the region where the slope of the pressure is 0 or more.
  • the evaluation value may be a pressure difference between two predetermined points in the time series data of pressure, or a rate of change in pressure between the predetermined two points. May be good.
  • FIG. 10B is a graph showing an enlarged area indicated by the arrow AR2 in the graph line F3 of FIG. 8A. As shown in FIG. 10B, when burrs occur in the molded product, the graph line F3 after the pressure holding release operation (time point X1) has a minimum (time point Xa, pressure P1) and a maximum (time point Xb, pressure). P2) appears.
  • the evaluation value may be the difference (P2-P1) between the pressure P1 at the minimum and the pressure P2 at the maximum.
  • the evaluation value is the rate of change in pressure ((P2-P1) / (Xb-Xa)) obtained by dividing the difference (P2-P1) by the time difference (Xb-Xa) between the minimum and the maximum. You may. That is, the evaluation value may be a value indicating a change in pressure after the pressure holding release operation.
  • the training data acquisition unit 31 acquires information input by the operator to the input unit 50.
  • the information input by the operator is, for example, burr information regarding the burr state of the molded product or first information stored in the molding information storage unit 33.
  • the training data acquisition unit 31 acquires the second information corresponding to the first information from the molding information storage unit 33 based on the first information input to the input unit 50. For example, when a lot number of a molding material is input to the input unit 50, the training data acquisition unit 31 acquires information on the physical properties of the molding material corresponding to the lot number from the molding information storage unit 33.
  • the burr information is, for example, information (for example, a dummy variable) that quantifies the presence or absence of burrs and the size of burrs (for example, the length and width of burrs).
  • the size of the burr the length of the burr itself may be quantified, or the degree of the size of the burr may be quantified.
  • the operator acquires burr information by actually inspecting the appearance of the molded product molded by the molding apparatus 20, and inputs the burr information to the input unit 50.
  • the learning calculation unit 32 generates a learned model that models the correlation between the evaluation value and the burr state of the molded product by performing a supervised machine learning calculation based on the training data.
  • a convolutional neural network (CCN) is used as the machine learning model, but other models may be used.
  • CCN convolutional neural network
  • it may be a regression tree model that is a model for grouping data.
  • the evaluation information includes the evaluation value (value related to the change in pressure after the holding pressure release operation), the second evaluation value (value related to temperature), the third evaluation value (value related to humidity, etc.), and molding information (value related to humidity, etc.). Includes molding material viscosity, etc.), mold clamping force and pressurization.
  • the learned model Tm1 generated by the learning calculation unit 32 is stored in the learned model storage unit 34.
  • the trained model Tm1 stored in the trained model storage unit 34 responds to the content of the training data. Will be updated as appropriate.
  • the learned model Tm1 is transmitted from the learning device 30 to the abnormality prediction device 40 described later, and is also stored in the learned model storage unit 45 of the abnormality prediction device 40.
  • the trained model Tm1 of the present embodiment may be stored and provided in an arbitrary storage medium such as a non-temporary computer-readable medium.
  • the method of generating the trained model Tm1 includes a training data acquisition step and a learning calculation step.
  • the training data acquisition unit 31 acquires the training data. For example, based on the time series data of the pressure detected by the pressure sensor 25 that detects the pressure of the molding material L1 in the cavity C1, the evaluation value regarding the change in the pressure of the molding material L1 after the pressure holding release step is acquired. In addition, the operator actually inspects the molded product for which the evaluation value has been acquired to acquire burr information.
  • the learning calculation unit 32 generates the trained model Tm1 based on the training data.
  • FIG. 11 is a block diagram showing a functional configuration of the abnormality prediction device 40 according to the present embodiment.
  • the abnormality prediction device 40 includes a data acquisition unit 41, an abnormality prediction unit 42, an output unit 43, a molding information storage unit 44, and a learned 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 an HDD.
  • the calculation unit executes the data acquisition process and the abnormality prediction process, which will be described later, based on the program stored in the storage unit.
  • the program of the present embodiment may be stored and provided on an arbitrary storage medium such as a non-transitory computer-readable medium.
  • the molding information storage unit 44 stores table-type molding information in which various first information and the second information are associated with each other.
  • the learned model Tm1 generated by the learning device 30 is stored in the learned model storage unit 45.
  • the molding information storage unit 44 and the learned model storage unit 45 may be realized by the same storage area as the molding information storage unit 33 and the learned model storage unit 34 of the learning device 30 among the computer devices, or may be different storage. It may be realized by the area. That is, the learning device 30 and the abnormality prediction device 40 may be configured to share the same molding information storage unit 33 and the learned model storage unit 34, or the learning device 30 and the abnormality prediction device 40 may be independently molded. It may be configured to have information storage units 33, 44 and trained model storage units 34, 45.
  • the data acquisition unit 41 executes a data acquisition process for acquiring evaluation information for performing abnormality prediction from each unit of the molding system 10.
  • the evaluation information is the same as the evaluation information acquired by the learning device 30 described above. That is, the evaluation information includes the evaluation value (value related to the change in pressure after the pressurization release operation), the second evaluation value (value related to temperature), the third evaluation value (value related to humidity, etc.), and the molding information (viscosity of the molding material). Etc.), including mold clamping force and pressurization.
  • the data acquisition unit 41 acquires an evaluation value regarding a change in pressure after the pressure holding release operation based on the pressure detected by the pressure sensor 25 and the pressurization sensor 227, respectively.
  • the evaluation value is the same value as the evaluation value acquired by the training data acquisition unit 31 of the learning device 30 described above. That is, the evaluation value is a value related to the change in pressure of the molding material L1 after the pressure holding release operation.
  • the abnormality prediction unit 42 executes an abnormality prediction process for acquiring prediction information for predicting the burr state of the molded product by inputting the evaluation information acquired by the data acquisition unit 41 into the trained model Tm1. .. In the abnormality prediction process, the information corresponding to the objective variable used when the trained model Tm1 is generated is output as the prediction information.
  • the prediction information is information including the probability of being predicted about the size of the burr of the molded product. More specifically, the prediction information includes, for example, the first probability that the burr length of the molded product is 1 mm or less, the second probability that it is longer than 1 mm and 5 mm or less, and the third probability that it is longer than 5 mm. ..
  • the trained model Tm1 outputs prediction information having a first probability of 10%, a second probability of 10%, and a third probability of 80%.
  • the prediction information may be information on the probability of being classified according to the presence or absence of burrs. That is, when the objective variable is the presence or absence of burrs on the molded product, the prediction information is the fourth probability without burrs and the fifth probability with burrs.
  • the output unit 43 determines the quality of the molded product based on the prediction information acquired by the abnormality prediction unit 42. For example, the output unit 43 determines the quality based on a predetermined threshold value and the above prediction information.
  • the predetermined threshold is, for example, the maximum length of burrs allowed. For example, when the maximum allowable length of burrs is 5 mm, it is determined that the molded product is defective (with burrs) when the third probability is the highest among the first to third probabilities described above. do.
  • the output unit 43 outputs the determination information regarding the quality of the molded product and the prediction information to the display unit 60 and the control unit 271. Judgment information and prediction information are displayed on the display unit 60. In particular, when it is determined that the molded product is defective, the determination information may be displayed in a highlighted color such as red on the display of the display unit 60, and an alert may be issued on the speaker.
  • the operation command of the control unit 271 causes the molding apparatus 20 that has molded the molded product determined to be defective to be stopped in a state where the mold portion 24 is open. It may be configured. In this case, the operator inspects the mold unit 24 based on the alert or the like by the display unit 60, and removes the foreign matter Fm1 (FIG. 6) as necessary.
  • the prediction information obtained by the abnormality prediction unit 42 may be displayed as it is on the display unit 60 without providing the output unit 43.
  • the operator may determine the quality of the molded product and the state of the molding apparatus 20 based on the prediction information displayed on the display unit 60.
  • the abnormality prediction method includes a data acquisition step and an abnormality prediction step.
  • the data acquisition unit 41 acquires the evaluation information. For example, based on the time series data of the pressure detected by the pressure sensor 25 that detects the pressure of the molding material L1 in the cavity C1, the evaluation value regarding the change in the pressure of the molding material L1 after the pressure holding release step is acquired.
  • the abnormality prediction unit 42 acquires prediction information for predicting the burr state of the molded product based on the evaluation information including the evaluation value.
  • the molding system 10 acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material L1 after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
  • the molding system 10 acquires prediction information by inputting the evaluation value into the trained model Tm1 in which the correlation between the evaluation value and the burr state of the molded product is machine-learned.
  • the explanatory variable of the trained model Tm1 includes an evaluation value
  • the objective variable of the trained model Tm1 is the state of burrs (for example, the presence or absence of burrs or the size of burrs) of the molded product. With this configuration, the burr state can be predicted more accurately even when the molding conditions vary.
  • the change in the pressure of the molding material L1 after the pressure holding release operation also depends on the temperature of the molding material L1.
  • the molding system 10 according to the present embodiment can be a trained model in which the above correlation is incorporated, and more accurately burrs. The state of can be predicted.
  • FIG. 12 is a block diagram schematically showing the molding system 11 according to the second embodiment.
  • the molding system 11 includes a plurality of molding devices 20, an abnormality prediction device 40a, an input unit 50, and a display unit 60.
  • the abnormality prediction device 40a of the molding system 11 acquires prediction information for predicting the burr state of the molded product based on the evaluation value and a predetermined reference value. That is, the molding system 11 is different from the molding system 10 according to the first embodiment in that the state of burrs is predicted by comparing the evaluation value and the reference value without using the trained model Tm1.
  • FIG. 13 is a block diagram showing a functional configuration of the abnormality prediction device 40a according to the present embodiment.
  • the abnormality prediction device 40a includes a data acquisition unit 41, an abnormality prediction 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 an HDD.
  • the data acquisition unit 41 acquires evaluation information in the same manner as the data acquisition unit 41 according to the first embodiment.
  • a plurality of reference values are stored in the reference value storage unit 46.
  • the reference value is a value generated based on the evaluation value obtained when the molded product is normally molded.
  • the reference value is, for example, a value obtained by adding a predetermined margin to the average value or the median value of a plurality of evaluation values obtained during normal molding of a plurality of molded products.
  • the predetermined margin is determined by the maximum length of burrs allowed in the molded product and the like.
  • a plurality of values are prepared for each of the second evaluation values, for each of the third evaluation values, or for each molding information.
  • the first reference value when the second evaluation value (value related to temperature) is less than the first value, the first reference value is used, when the second evaluation value is equal to or more than the first value and less than the second value, the second reference value and the second evaluation value are the second.
  • the third reference value may be used.
  • the reference value storage unit 46 stores, for example, table-type information in which the second evaluation value, the third evaluation value, and the molding information are associated with a plurality of reference values.
  • the abnormality prediction unit 42a has a reference value (corresponding from a plurality of reference values stored in the reference value storage unit 46, based on the second evaluation value, the third evaluation value, and the molding information acquired by the data acquisition unit 41. Hereinafter, it is referred to as “attention reference value”). Then, the abnormality prediction unit 42a compares the evaluation value acquired by the data acquisition unit 41 (that is, the evaluation value acquired at the time of molding the molded product to be predicted) with the attention reference value to predict the prediction information. To get.
  • the prediction information is, for example, the difference or ratio between the evaluation value and the attention reference value.
  • the evaluation value for example, the slope of pressure
  • the evaluation value tends to be larger than the reference value of interest. Therefore, when the evaluation value is larger than the attention reference value, it is predicted that burrs are generated in the molded product to be predicted. Therefore, the prediction information indicates whether or not burrs are generated in the molded product to be predicted.
  • FIG. 14 is a graph schematically explaining the comparison between the evaluation value and the attention reference value in the abnormality prediction unit 42a.
  • the vertical axis is the slope of pressure (dP / dt), and the horizontal axis is time t.
  • FIG. 14 also shows the attention reference value Th1.
  • the graph lines F34 and F35 are time-series data of the pressure gradient acquired by the data acquisition unit 41 based on the time-series data of the pressure obtained by the pressure sensor 25 when the molded product to be predicted is molded, respectively. be.
  • the evaluation values are dP4 and dP5 in the example of FIG.
  • the evaluation value dP4 is larger than the attention reference value Th1
  • the evaluation value dP5 is smaller than the attention reference value Th1
  • no burrs are generated in the molded product from which the graph line F35 is acquired (or even if burrs are generated, they are within the permissible range. Yes) is expected.
  • the output unit 43a determines whether or not the burr state of the molded product is good or bad based on the prediction information acquired by the abnormality prediction unit 42a. For example, when the prediction information is the difference between the evaluation value and the attention reference value (evaluation value-attention reference value), the output unit 43a has a poor burr state of the molded product when the prediction information is a positive value. Judged as (abnormal).
  • the output unit 43a outputs the determination information regarding the quality of the burr of the molded product and the prediction information to the display unit 60 and the control unit 271.
  • the display unit 60 and the control unit 271 perform the same operations as in the first embodiment based on the determination information and the prediction information.
  • the prediction information obtained by the abnormality prediction unit 42a may be displayed as it is on the display unit 60 without providing the output unit 43a.
  • the operator may determine the quality of the molded product and the state of the molding apparatus 20 based on the prediction information displayed on the display unit 60.
  • the state of burrs can be easily predicted by comparing the reference value and the evaluation value.
  • Molding system 20 Molding equipment 21 Bed 22 Injection part 221 Hopper 222 Cylinder 223 Screw 224 Nozzle 225 Ball screw 226 Motor 227 Pressure sensor 228 Heater 23 Type tightening part 231 Fixed plate 232 Movable plate 233 Tie bar 234 Ball screw 235 Support plate 236 Mold tightening force sensor 237 Motor 24 Mold part 241 Mold 241a Exposed surface 242 Mold 243 Flow path 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 Learned model storage unit 40, 40a Abnormality prediction device 41 Data acquisition unit 42, 42a Abnormality prediction unit 43, 43a Output unit 44 Molding information storage unit 45 Learned model storage unit 46 Reference value storage unit 50 Input Part 60 Display part C1 Cavity C1a Area C2 Gap L1 Molding material Fm1 Foreign matter

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Mechanical Engineering (AREA)
  • Injection Moulding Of Plastics Or The Like (AREA)
  • Moulds For Moulding Plastics Or The Like (AREA)

Abstract

A molding system (10) comprises a molding device (20) and an irregularity prediction device (40). The irregularity prediction device (40) is configured to acquire an evaluation value pertaining to a change of pressure of a molding material after a holding–pressure–releasing operation of the molding device (20) on the basis of time series data of the pressure of the molding material in a cavity of a die of the molding device (20), and, on the basis of the evaluation value, to acquire prediction information for predicting the burr condition of a molded product.

Description

成形システム、異常予測装置、異常予測方法、プログラム及び学習済みモデルMolding system, anomaly prediction device, anomaly prediction method, program and trained model
　本開示は、成形システム、異常予測装置、異常予測方法、プログラム及び学習済みモデルに関する。 This disclosure relates to molding systems, anomaly predictors, anomaly prediction methods, programs and trained models.
　複数の金型の間に形成されるキャビティへ成形材料を溶融させた融液を供給して成型品を成形する射出成形装置が知られている。成型品の不良として、例えばバリが挙げられる。バリが発生すると、成型品の外観を損ねるだけでなく、成型品の性能に影響を及ぼす場合もある。このため、現在は目視による成型品の検査を行っている。 An injection molding apparatus is known in which a melt obtained by melting a molding material is supplied to a cavity formed between a plurality of molds to mold a molded product. Examples of defective molded products include burrs. The occurrence of burrs not only spoils the appearance of the molded product, but may also affect the performance of the molded product. For this reason, we are currently visually inspecting molded products.
　特許文献１には、オペレータがバリの発生に関して現在の射出条件の適否を判断するために、圧力センサにより検出された金型内の圧力の経時変化を示す曲線と、バリが吹く目安となる圧力であるバリ吹き限界曲線とを表示部に描画する技術が開示されている。バリ吹き限界曲線は、実験によって各種の値を取得することで描画することができる曲線である。 In Patent Document 1, in order for the operator to judge the suitability of the current injection conditions with respect to the occurrence of burrs, a curve showing the time-dependent change of the pressure in the mold detected by the pressure sensor and the pressure as a guide for the burrs to blow are described. A technique for drawing a burr blowing limit curve on a display unit is disclosed. The burr blowing limit curve is a curve that can be drawn by acquiring various values by experiments.
日本国特開２０１９－１３９３３公報Japanese Patent Application Laid-Open No. 2019-13933
　特許文献１に係るバリ吹き限界曲線は、成形条件ごとに異なる曲線となる。成形条件には、金型の形状、成形材料の種類・粘度・温度、成形材料（融液）に印加する圧力、成形環境（室温、湿度）などが含まれ、多岐にわたる。このため、特許文献１に係る技術に基づいてバリの発生を判断するためには、成形条件ごとにバリ吹き限界曲線を用意する必要がある。また、実際の成形条件が、バリ吹き限界曲線の基礎となる条件から外れると、バリが発生しているのにバリ吹き限界曲線よりも実際の圧力が低くなる等して、バリの発生を正確に判断できない場合がある。 The burr blowing limit curve according to Patent Document 1 is a different curve depending on the molding conditions. Molding conditions include the shape of the mold, the type / viscosity / temperature of the molding material, the pressure applied to the molding material (melt), the molding environment (room temperature, humidity), and the like. Therefore, in order to determine the occurrence of burrs based on the technique according to Patent Document 1, it is necessary to prepare a burr blowing limit curve for each molding condition. In addition, if the actual molding conditions deviate from the conditions that form the basis of the burr blowing limit curve, the actual pressure will be lower than the burr blowing limit curve even though burrs have occurred, and the occurrence of burrs will be accurate. It may not be possible to judge.
　そこで、本開示は、より正確にバリの状態を予測するための成形システム、異常予測装置、異常予測方法、プログラム及び学習済みモデルを提供することを目的とする。 Therefore, an object of the present disclosure is to provide a molding system, an abnormality prediction device, an abnormality prediction method, a program, and a trained model for more accurately predicting the state of burrs.
（１）　本開示は、成形装置と、前記成形装置により成形される成型品の異常を予測する異常予測装置と、を備える成形システムであって、前記成形装置は、複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め部と、前記複数の金型を締め付けることで前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填動作と、前記キャビティに充填された前記成形材料の圧力を保持する保圧動作と、前記成形材料の圧力の保持を解除する保圧解除動作と、を行う射出部と、前記キャビティ内の前記成形材料の圧力を検出する圧力センサと、を有し、前記異常予測装置は、前記圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除動作後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得部と、前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測部と、を有する、成形システムを提供する。 (1) The present disclosure is a molding system including a molding device and an abnormality predicting device for predicting an abnormality of a molded product molded by the molding device, and the molding device combines a plurality of molds. In this state, a mold clamping portion for tightening the plurality of molds, a filling operation for filling a cavity formed between the plurality of molds by tightening the plurality of molds with a molten molding material, and the above-mentioned The injection portion that performs the pressure holding operation for holding the pressure of the molding material filled in the cavity and the pressure holding release operation for releasing the pressure holding of the molding material, and the pressure of the molding material in the cavity. It has a pressure sensor to detect, and the abnormality predictor determines an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation based on the time series data of the pressure detected by the pressure sensor. Provided is a molding system having a data acquisition unit to be acquired and an abnormality prediction unit for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value.
　発明者らは、鋭意研究の結果、保圧解除動作後のキャビティ内の圧力が、正常時（バリ非発生時）には単調減少する一方で、バリ発生時には圧力が一瞬だけ増加、非減少、又は減少の程度が小さくなる（圧力の傾きが一瞬だけ大きくなる）ことを発見した。すなわち、保圧解除動作後の成形材料の圧力の変化に着目すれば、バリの状態を予測することができることを発見した。そこで、本開示に係る成形システムは、保圧解除動作後の成形材料の圧力の変化に基づいて、バリの状態を予測するための予測情報を取得する。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる１つの圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 As a result of diligent research, the inventors monotonically decreased the pressure in the cavity after the pressure release operation during normal times (when burrs did not occur), while the pressure increased or decreased momentarily when burrs occurred. Alternatively, it was discovered that the degree of decrease becomes smaller (the pressure gradient becomes larger for a moment). That is, it was discovered that the state of burrs can be predicted by paying attention to the change in the pressure of the molding material after the pressure holding release operation. Therefore, the molding system according to the present disclosure acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
（２）　本開示は、上記（１）に記載の成形システムであって、前記異常予測部は、前記評価値と前記成型品のバリの状態との相関関係を機械学習させた学習済みモデルへ前記評価値を入力することで、前記予測情報を取得し、前記学習済みモデルの説明変数は、前記評価値を含み、前記学習済みモデルの目的変数は、前記成型品のバリの状態であり、前記バリの状態は、前記バリの有無、又は前記バリの大きさを含む、成形システムを提供してもよい。 (2) The present disclosure is the molding system described in (1) above, and the abnormality prediction unit uses a trained model in which the correlation between the evaluation value and the burr state of the molded product is machine-learned. By inputting the evaluation value, the prediction information is acquired, the explanatory variable of the trained model includes the evaluation value, and the objective variable of the trained model is the burr state of the molded product. The burr state may provide a molding system including the presence or absence of the burr or the size of the burr.
　このように構成することで、成形条件にばらつきがある状態であっても、より正確にバリの状態を予測することができる。 With this configuration, the burr state can be predicted more accurately even if the molding conditions vary.
（３）　本開示は、上記（２）に記載の成形システムであって、前記成形装置は、前記成形材料の温度を検出する温度センサをさらに有し、前記学習済みモデルの前記説明変数は、前記温度センサにより検出された温度に関する第２評価値を含み、前記異常予測部は、前記学習済みモデルへ前記評価値及び前記第２評価値を入力することで、前記予測情報を取得する、成形システムを提供してもよい。 (3) The present disclosure is the molding system according to (2) above, wherein the molding apparatus further includes a temperature sensor for detecting the temperature of the molding material, and the explanatory variables of the trained model are: Molding that includes a second evaluation value related to the temperature detected by the temperature sensor, and the abnormality prediction unit acquires the prediction information by inputting the evaluation value and the second evaluation value into the trained model. A system may be provided.
　保圧解除動作後の成形材料の圧力の変化は、成形材料の温度にも依存する。成形材料の温度が高いほど、成形材料の粘度が低く、成形材料が流動しやすくなる。このため、成形材料の温度が高いほど、バリ発生時の圧力の変化が短時間のうちに急峻に生じやすくなる。成形材料の温度に関する第２評価値を説明変数に含むことで、上記の相関関係が組み込まれた学習済みモデルとすることが可能となり、より正確にバリの状態を予測することができる。 The change in the pressure of the molding material after the pressure holding release operation also depends on the temperature of the molding material. The higher the temperature of the molding material, the lower the viscosity of the molding material and the easier it is for the molding material to flow. Therefore, the higher the temperature of the molding material, the more likely it is that the pressure change at the time of burr generation will occur sharply within a short period of time. By including the second evaluation value regarding the temperature of the molding material in the explanatory variable, it becomes possible to obtain a trained model in which the above correlation is incorporated, and the state of burrs can be predicted more accurately.
（４）　本開示は、上記（１）に記載の成形システムであって、前記異常予測部は、前記成型品の正常成形時に取得される前記評価値に基づいて生成される基準値よりも、予測対象となる前記成型品の成形時に取得される前記評価値が大きくなる場合に、前記予測対象となる前記成型品にバリが発生していることを示す前記予測情報を取得する、成形システムを提供してもよい。このように構成することで、基準値と評価値との比較により、バリの状態を容易に予測することができる。 (4) The present disclosure is the molding system according to (1) above, and the abnormality prediction unit is more than a reference value generated based on the evaluation value acquired at the time of normal molding of the molded product. A molding system that acquires the prediction information indicating that burrs are generated in the molded product to be predicted when the evaluation value acquired at the time of molding the molded product to be predicted becomes large. May be provided. With this configuration, the state of burrs can be easily predicted by comparing the reference value and the evaluation value.
（５）　本開示は、上記（１）～（４）のいずれかに記載の成形システムであって、前記データ取得部は、前記保圧解除動作後の前記時系列データに基づいて得られる圧力の傾きの第２時系列データに基づいて、前記評価値を取得し、前記評価値は、前記第２時系列データにおける前記傾きの最大値、又は前記第２時系列データのうち前記傾きが０以上となる領域の幅である、成形システムを提供してもよい。このように構成することで、時系列データから容易に評価値を取得することができる。 (5) The present disclosure is the molding system according to any one of (1) to (4) above, and the data acquisition unit is a pressure obtained based on the time series data after the pressure holding release operation. The evaluation value is acquired based on the second time-series data of the inclination of, and the evaluation value is the maximum value of the inclination in the second time-series data or the inclination of 0 in the second time-series data. You may provide a molding system which is the width of the above area. With this configuration, the evaluation value can be easily obtained from the time series data.
（６）　本開示は、上記（１）～（４）のいずれかに記載の成形システムであって、前記データ取得部は、前記保圧解除動作後の前記時系列データの極小における圧力と極大における圧力との差、又は前記極小における圧力から前記極大における圧力への変化率を、前記評価値として取得する、成形システムを提供してもよい。このように構成することで、時系列データから容易に評価値を取得することができる。 (6) The present disclosure is the molding system according to any one of (1) to (4) above, and the data acquisition unit maximizes the pressure at the minimum of the time series data after the pressure holding release operation. A molding system may be provided that acquires, as the evaluation value, the difference from the pressure at the above, or the rate of change from the pressure at the minimum to the pressure at the maximum. With this configuration, the evaluation value can be easily obtained from the time series data.
（７）　本開示は、成形装置により成形される成型品の異常を予測する異常予測装置であって、前記成形装置は、複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め部と、前記複数の金型を締め付けることで前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填動作と、前記キャビティに充填された前記成形材料の圧力を保持する保圧動作と、前記成形材料の圧力の保持を解除する保圧解除動作と、を行う射出部と、前記キャビティ内の前記成形材料の圧力を検出する圧力センサと、を有し、前記異常予測装置は、前記圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除動作後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得部と、前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測部と、を備える、異常予測装置を提供する。 (7) The present disclosure is an abnormality prediction device for predicting an abnormality in a molded product molded by a molding device, wherein the molding device is a mold for tightening the plurality of molds in a state where a plurality of molds are combined. The filling operation of filling the molded material in a molten state into the tightening portion and the cavity formed between the plurality of molds by tightening the plurality of molds, and the pressure of the molding material filled in the cavity. It has an injection unit that performs a holding pressure holding operation and a pressure holding release operation that releases the pressure holding of the molding material, and a pressure sensor that detects the pressure of the molding material in the cavity. The abnormality prediction device includes a data acquisition unit that acquires an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation based on the time-series data of the pressure detected by the pressure sensor, and the evaluation value. Based on this, the present invention provides an abnormality prediction device including an abnormality prediction unit for acquiring prediction information for predicting a burr state of the molded product.
　本開示に係る異常予測装置は、保圧解除動作後の成形材料の圧力の変化に基づいて、バリの状態を予測するための予測情報を取得する。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる１つの圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 The abnormality prediction device according to the present disclosure acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
（８）　本開示は、複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測する異常予測方法であって、前記キャビティ内の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除工程後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得工程と、前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測工程と、を備える、異常予測方法を提供する。 (8) In the present disclosure, in a state where a plurality of molds are combined, a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step. Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material. It is an abnormality prediction method for predicting an abnormality of a molded product to be formed, and is after the pressure holding release step based on time-series data of pressure detected by a pressure sensor that detects the pressure of the molding material in the cavity. It includes a data acquisition step of acquiring an evaluation value regarding a change in pressure of the molding material, and an abnormality prediction step of acquiring prediction information for predicting a burr state of the molded product based on the evaluation value. An abnormality prediction method is provided.
　本開示に係る異常予測方法は、保圧解除動作後の成形材料の圧力の変化に基づいて、バリの状態を予測するための予測情報を取得する。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる１つの圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 The abnormality prediction method according to the present disclosure acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
（９）　本開示は、複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測するためのプログラムであって、前記キャビティ内の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除工程後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得処理と、前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測処理と、をコンピュータ装置に実行させる、プログラムを提供する。 (9) In the present disclosure, in a state where a plurality of molds are combined, a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step. Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material. It is a program for predicting an abnormality of a molded product to be formed, and is after the pressure holding release step based on time-series data of pressure detected by a pressure sensor that detects the pressure of the molding material in the cavity. A computer device includes a data acquisition process for acquiring an evaluation value regarding a change in pressure of the molding material, and an abnormality prediction process for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value. Provide a program to be executed by.
　本開示に係るプログラムを実行すれば、保圧解除動作後の成形材料の圧力の変化に基づいて、バリの状態を予測するための予測情報を取得することができる。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる１つの圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 By executing the program according to the present disclosure, it is possible to acquire prediction information for predicting the state of burrs based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
（１０）　本開示は、複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測するための学習済みモデルであって、説明変数は、前記キャビティ内の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて取得される、前記保圧解除動作後の前記成形材料の圧力の変化に関する評価値を含み、目的関数は、前記成型品のバリの状態であり、前記バリの状態は、前記バリの有無、又は前記バリの大きさを含む、学習済みモデルを提供する。 (10) In the present disclosure, in a state where a plurality of molds are combined, a mold clamping step of tightening the plurality of molds and a state of being melted into a cavity formed between the plurality of molds by the mold clamping step. Molding by a molding method including a filling step of filling the molding material, a pressure holding step of holding the pressure of the molding material filled in the cavity, and a pressure holding release step of releasing the pressure holding of the molding material. It is a trained model for predicting the abnormality of the molded product to be made, and the explanatory variables are acquired based on the time-series data of the pressure detected by the pressure sensor that detects the pressure of the molding material in the cavity. The objective function is the burr state of the molded product, and the burr state is the presence or absence of the burr or the burr. Provides a trained model, including the size of.
　本開示に係る学習済みモデルは、説明関数を保圧解除動作後の成形材料の圧力の変化に関する評価値とし、目的関数を成型品のバリの状態としているため、学習済みモデルへ評価値を入力すれば、予測したバリの状態が出力される。これにより、保圧解除動作後の成形材料の圧力の変化に基づいて、バリの状態を予測することができる。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 In the trained model according to the present disclosure, since the explanatory function is the evaluation value regarding the change in the pressure of the molding material after the pressure release operation and the objective function is the burr state of the molded product, the evaluation value is input to the trained model. Then, the predicted burr state is output. Thereby, the state of burrs can be predicted based on the change in the pressure of the molding material after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from the pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
　本開示によれば、より正確にバリの状態を予測することができる。 According to this disclosure, the state of burrs can be predicted more accurately.
図１は、第１実施形態に係る成形システムを模式的に示すブロック図である。FIG. 1 is a block diagram schematically showing a molding system according to the first embodiment. 図２は、図１に係る成形装置を概念的に示す説明図である。FIG. 2 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1. 図３は、図１に係る成形装置を概念的に示す説明図である。FIG. 3 is an explanatory diagram conceptually showing the molding apparatus according to FIG. 1. 図４は、充填工程の終了時の様子を模式的に示す金型部の断面図である。FIG. 4 is a cross-sectional view of a mold portion schematically showing a state at the end of the filling process. 図５（ａ）及び図５（ｂ）は、圧力及び圧力の傾きの時系列データを示すグラフの一例である。5 (a) and 5 (b) are examples of graphs showing time-series data of pressure and pressure gradient. 図６は、成型品にバリが発生する場合の、充填工程の終了時の様子を模式的に示す金型部の断面図である。FIG. 6 is a cross-sectional view of a mold portion schematically showing a state at the end of a filling process when burrs are generated in a molded product. 図７は、成型品にバリが発生する場合の、保圧解除工程の様子を模式的に示す断面図である。FIG. 7 is a cross-sectional view schematically showing a state of the pressure holding release step when burrs are generated in the molded product. 図８（ａ）及び図８（ｂ）は、成型品にバリが発生する場合の、圧力及び圧力の傾きの時系列データを示すグラフの一例である。8 (a) and 8 (b) are examples of graphs showing time-series data of pressure and pressure gradient when burrs are generated in the molded product. 図９は、第１実施形態に係る学習装置の機能構成を示すブロック図である。FIG. 9 is a block diagram showing a functional configuration of the learning device according to the first embodiment. 図１０（ａ）及び図１０（ｂ）は、第１実施形態に係る評価値について説明する説明図である。10 (a) and 10 (b) are explanatory views for explaining the evaluation values according to the first embodiment. 図１１は、第１実施形態に係る異常予測装置の機能構成を示すブロック図である。FIG. 11 is a block diagram showing a functional configuration of the abnormality prediction device according to the first embodiment. 図１２は、第２実施形態に係る成形システムを模式的に示すブロック図である。FIG. 12 is a block diagram schematically showing the molding system according to the second embodiment. 図１３は、第２実施形態に係る異常予測装置の機能構成を示すブロック図である。FIG. 13 is a block diagram showing a functional configuration of the abnormality prediction device according to the second embodiment. 図１４は、第２実施形態に係る評価値と注目基準値との比較を模式的に説明するグラフである。FIG. 14 is a graph schematically explaining the comparison between the evaluation value and the attention reference value according to the second embodiment.
　＜第１実施形態＞
　以下、本開示の第１実施形態を、図面を参照して説明する。 <First Embodiment>
Hereinafter, the first embodiment of the present disclosure will be described with reference to the drawings.
　＜成形システムの全体構成＞
　図１は、第１実施形態に係る成形システム１０を模式的に示すブロック図である。成形システム１０は、複数の成形装置２０と、学習装置３０と、異常予測装置４０と、入力部５０と、表示部６０とを備える。 <Overall configuration of molding system>
FIG. 1 is a block diagram schematically showing a molding system 10 according to the first embodiment. The molding system 10 includes a plurality of molding devices 20, a learning device 30, an abnormality prediction device 40, an input unit 50, and a display unit 60.
　成形装置２０、学習装置３０、異常予測装置４０、入力部５０及び表示部６０は、それぞれ無線又は有線により通信可能に設けられている。学習装置３０及び異常予測装置４０は、演算部（例えば、ＣＰＵ、ＧＰＵ等）と、記憶部（例えば、ＨＤＤ、ＳＳＤ等）とを有する情報処理装置（コンピュータ装置）により構成されている。学習装置３０及び異常予測装置４０は、同一の情報処理装置により構成されてもよいし、別々の情報処理装置により構成されてもよい。 The molding device 20, the learning device 30, the abnormality prediction device 40, the input unit 50, and the display unit 60 are each provided so as to be able to communicate wirelessly or by wire. The learning device 30 and the abnormality prediction device 40 are composed of an information processing device (computer device) having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.). The learning device 30 and the abnormality prediction device 40 may be configured by the same information processing device or may be configured by separate information processing devices.
　本実施形態では複数の成形装置２０が１個の学習装置３０及び１個の異常予測装置４０に接続され、学習装置３０及び異常予測装置４０は複数の成形装置２０から送信される各種のデータに基づいて学習及び異常予測を行う。なお、成形装置２０は学習装置３０及び異常予測装置４０と１対１で対応していてもよい。すなわち、成形システム１０において、成形装置２０は１個であってもよいし、学習装置３０及び異常予測装置４０は複数備えられていてもよい。 In the present embodiment, a plurality of molding devices 20 are connected to one learning device 30 and one abnormality prediction device 40, and the learning device 30 and the abnormality prediction device 40 are used for various data transmitted from the plurality of molding devices 20. Based on this, learning and abnormality prediction are performed. The molding device 20 may have a one-to-one correspondence with the learning device 30 and the abnormality prediction device 40. That is, in the molding system 10, the molding device 20 may be one, or a plurality of learning devices 30 and abnormality prediction devices 40 may be provided.
　入力部５０は、例えばキーボードやマウスであり、オペレータからの各種の入力を受付ける。表示部６０は、例えばディスプレイやスピーカであり、成形システム１０における各種の情報を表示する。入力部５０及び表示部６０は、例えばタッチパネルのように一体となっていてもよい。また、入力部５０及び表示部６０は、携帯型の端末装置として、成形装置２０、学習装置３０及び異常予測装置４０から離れた場所に移動可能に設けられていてもよい。 The input unit 50 is, for example, a keyboard or a mouse, and receives various inputs from the operator. The display unit 60 is, for example, a display or a speaker, and displays various information in the molding system 10. The input unit 50 and the display unit 60 may be integrated, for example, a touch panel. Further, the input unit 50 and the display unit 60 may be provided as a portable terminal device so as to be movable to a place away from the molding device 20, the learning device 30, and the abnormality prediction device 40.
　＜成形装置の概略構成＞
　図２及び図３は、成形装置２０を概念的に示す説明図である。図２及び図３において、断面として示す部分にはハッチングを付す。成形装置２０は、ベッド２１と、射出部２２と、型締め部２３と、金型部２４と、圧力センサ２５と、温度センサ２６と、制御盤２７とを有する。図２は、金型部２４が開放されている状態の成形装置２０を示しており、図３は、金型部２４が組み合わされている状態の成形装置２０を示している。成形装置２０は、型締め式の射出成形を行う装置である。 <Outline configuration of molding equipment>
2 and 3 are explanatory views conceptually showing the molding apparatus 20. In FIGS. 2 and 3, the portion shown as a cross section is hatched. The molding apparatus 20 includes a bed 21, an injection portion 22, a mold clamping portion 23, a mold portion 24, a pressure sensor 25, a temperature sensor 26, and a control panel 27. FIG. 2 shows a molding apparatus 20 in a state where the mold portion 24 is open, and FIG. 3 shows a molding apparatus 20 in a state where the mold portion 24 is combined. The molding apparatus 20 is an apparatus for performing mold-fastening injection molding.
　制御盤２７は、制御部２７１と、通信部２７２とを有する。制御部２７１は成形装置２０の各駆動部（モータ２３７等）と電気的に接続し、当該各駆動部へ動作指令を出力する。また、制御部２７１は成形装置２０の各センサ（圧力センサ２５等）と電気的に接続し、当該各センサにより検出された信号が制御部２７１へ入力される。制御部２７１は、演算部（例えば、ＣＰＵ、ＧＰＵ等）と、記憶部（例えば、ＨＤＤ、ＳＳＤ等）とを有する情報処理装置により構成されている。 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 apparatus 20 and outputs an operation command to each drive unit. Further, the control unit 271 is electrically connected to each sensor (pressure sensor 25 or the like) of the molding apparatus 20, and the signal detected by each sensor is input to the control unit 271. The control unit 271 is composed of an information processing device having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.).
　通信部２７２は、成形システム１０の他の各部（学習装置３０等）と通信を行う。通信部２７２は、例えば、当該各センサにより検出された信号を学習装置３０又は異常予測装置４０へ送信する。また、通信部２７２は、後述の判定情報及び予測情報を異常予測装置４０から受信する。 The communication unit 272 communicates with each other unit (learning device 30, etc.) of the molding system 10. The communication unit 272 transmits, for example, the signal detected by each of the sensors to the learning device 30 or the abnormality prediction device 40. Further, the communication unit 272 receives the determination information and the prediction information described later from the abnormality prediction device 40.
　型締め部２３は、固定盤２３１と、可動盤２３２と、タイバー２３３と、ボールねじ２３４と、支持盤２３５と、型締め力センサ２３６と、モータ２３７とを有する。固定盤２３１及び支持盤２３５は、ベッド２１に固定されている。支持盤２３５はボールねじ２３４を支持している。ボールねじ２３４はモータ２３７と接続している。制御部２７１の動作指令によりモータ２３７が回転されると、ボールねじ２３４は移動する。ボールねじ２３４のうち、モータ２３７と接続されている端部とは反対側の端部には、可動盤２３２が固定されている。 The mold clamping portion 23 includes a fixed plate 231, a movable plate 232, a tie bar 233, a ball screw 234, a support plate 235, a mold clamping force sensor 236, and a motor 237. The fixing plate 231 and the supporting plate 235 are fixed to the bed 21. The support board 235 supports the ball screw 234. The ball screw 234 is connected to the motor 237. When the motor 237 is rotated by the operation command of the control unit 271, the ball screw 234 moves. A movable platen 232 is fixed to the end of the ball screw 234 opposite to the end connected to the motor 237.
　ここで、成形装置２０において、ボールねじ２３４が移動する方向を「軸方向」と称する。ボールねじ２３４に対してモータ２３７が位置する側を軸方向の「一方側」と称し、ボールねじ２３４に対して可動盤２３２が位置する側を軸方向の「他方側」と称する。 Here, in the molding apparatus 20, the direction in which the ball screw 234 moves is referred to as an "axial direction". The side where the motor 237 is located with respect to the ball screw 234 is referred to as the "one side" in the axial direction, and the side where the movable platen 232 is located with respect to the ball screw 234 is referred to as the "other side" in the axial direction.
　可動盤２３２は、ボールねじ２３４の移動に伴って、軸方向に移動する。可動盤２３２には、軸方向に貫通している貫通孔２３２ａが形成されている。タイバー２３３は、軸方向一方側の端部が支持盤２３５に固定され、軸方向他方側の端部が固定盤２３１に固定されている。タイバー２３３は可動盤２３２の貫通孔２３２ａに挿入されている。これにより、タイバー２３３は、可動盤２３２の軸方向の移動を案内する。 The movable board 232 moves in the axial direction as the ball screw 234 moves. The movable platen 232 is formed with a through hole 232a penetrating in the axial direction. The end of the tie bar 233 on one side in the axial direction is fixed to the support plate 235, and the end on the other side in the axial direction is fixed to the fixing plate 231. The tie bar 233 is inserted into the through hole 232a of the movable platen 232. As a result, the tie bar 233 guides the axial movement of the movable platen 232.
　型締め力センサ２３６は、ボールねじ２３４から支持盤２３５に加えられる圧力（型締め力の反力）を検出する。型締め力センサ２３６は、圧力に関する検出信号を、制御部２７１に出力する。なお、型締め力センサ２３６は、後述の金型部２４における型締め力を検出できる位置であれば、他の位置に設置されていてもよい。固定盤２３１には、軸方向他方側に径が広がる貫通孔２３１ａが形成されている。貫通孔２３１ａには、後述のシリンダ２２２が挿入される。 The mold clamping force sensor 236 detects the pressure (reaction force of the mold clamping force) applied to the support plate 235 from the ball screw 234. The mold clamping force sensor 236 outputs a detection signal regarding pressure to the control unit 271. The mold clamping force sensor 236 may be installed at another position as long as it can detect the mold clamping force in the mold portion 24 described later. The fixing plate 231 is formed with a through hole 231a having a diameter widened on the other side in the axial direction. A cylinder 222, which will be described later, is inserted into the through hole 231a.
　金型部２４は、複数の金型２４１、２４２を有する。金型２４１は、可動盤２３２に固定されている。ボールねじ２３４が軸方向に移動すると、可動盤２３２とともに金型２４１も軸方向に移動する。すなわち、金型２４１は可動金型である。金型２４２は、固定盤２３１に固定されている。すなわち、金型２４２は固定金型である。金型２４２には、流路２４３が形成されている。 The mold unit 24 has a plurality of molds 241 and 242. The mold 241 is fixed to the movable platen 232. When the ball screw 234 moves in the axial direction, the mold 241 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 fixing plate 231. That is, the mold 242 is a fixed mold. A flow path 243 is formed in the mold 242.
　図３に示すように、ボールねじ２３４により金型２４１が軸方向他方側に移動し、金型２４１が金型２４２に接触すると（すなわち、複数の金型２４１、２４２が組み合わされると）、金型２４１、２４２の間にはキャビティＣ１（空間）が形成される。 As shown in FIG. 3, when the mold 241 is moved to the other side in the axial direction by the ball screw 234 and the mold 241 comes into contact with the mold 242 (that is, when a plurality of molds 241 and 242 are combined), the mold Cavity C1 (space) is formed between the molds 241 and 242.
　射出部２２は、ホッパ２２１と、シリンダ２２２と、スクリュ２２３と、ボールねじ２２５と、モータ２２６と、与圧センサ２２７と、ヒータ２２８とを有する。ホッパ２２１は、シリンダ２２２と接続しており、シリンダ２２２内に成形材料を供給する。シリンダ２２２は軸方向に延びる中空の円筒形状を有する部材である。シリンダ２２２の軸方向一方側の端部は、径方向一方側の最端に近づくにつれて径が狭くなっており、当該最端にはノズル２２４が設けられている。ノズル２２４は、金型２４２の流路２４３と接続している。 The injection unit 22 includes a hopper 221, a cylinder 222, a screw 223, a ball screw 225, a motor 226, a pressurization sensor 227, and a heater 228. The hopper 221 is connected to the cylinder 222 and supplies the 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 one end of the cylinder 222 on one side in the axial direction becomes narrower as it approaches the end on one side in the radial direction, and a nozzle 224 is provided at the end. The nozzle 224 is connected to the flow path 243 of the mold 242.
　スクリュ２２３は、シリンダ２２２の軸方向他方側の端部からシリンダ２２２内に挿入されている。スクリュ２２３の軸方向他方側にはボールねじ２２５が接続されており、ボールねじ２２５の軸方向他方側にはモータ２２６が接続されている。制御部２７１の動作指令によりモータ２２６が回転すると、ボールねじ２２５は軸方向に移動する。これに伴いスクリュ２２３も軸方向に移動する。このとき、スクリュ２２３は、軸方向を中心軸とする周方向に回転する。 The screw 223 is inserted into the cylinder 222 from the end on the other side in the axial direction of the cylinder 222. A ball screw 225 is connected to the other side of the screw 223 in the axial direction, and a motor 226 is connected to the other side of the ball screw 225 in the axial direction. When the motor 226 is rotated by the operation command of the control unit 271, the ball screw 225 moves in the axial direction. Along with this, the screw 223 also moves in the axial direction. At this time, the screw 223 rotates in the circumferential direction with the axial direction as the central axis.
　与圧センサ２２７は、ボールねじ２２５からモータ２２６へ加えられる圧力（スクリュ２２３の押込み力の反力）を検出する。与圧センサ２２７は、圧力に関する検出信号を、制御部２７１に出力する。なお、与圧センサ２２７は、スクリュ２２３の押込み力を検出できる位置であれば、他の位置に設置されていてもよい。 The pressurization sensor 227 detects the pressure applied from the ball screw 225 to the motor 226 (the reaction force of the pushing force of the screw 223). The pressurization sensor 227 outputs a detection signal related to pressure to the control unit 271. The pressurization sensor 227 may be installed at another position as long as it can detect the pushing force of the screw 223.
　ヒータ２２８は、例えば抵抗線をコイル状に巻回した抵抗加熱ヒータである。ヒータ２２８は、制御部２７１の動作指令により当該抵抗線へ電流が流されることで、抵抗熱によりシリンダ２２２内を加熱する。 The heater 228 is, for example, a resistance heating heater in which a resistance wire is wound in a coil shape. The heater 228 heats the inside of the cylinder 222 by the resistance heat when a current is passed through the resistance wire by the operation command of the control unit 271.
　圧力センサ２５は、金型２４１、２４２のうちキャビティＣ１に面する領域に設置されている。圧力センサ２５は、キャビティＣ１内の圧力を検出する。特に、圧力センサ２５は、キャビティＣ１内に供給された成形材料（溶融状態もしくは固化状態、又は溶融状態と固化状態とが混在した状態）の圧力を検出する。圧力センサ２５は、圧力に関する検出信号を制御部２７１に出力する。本実施形態において、圧力センサ２５は金型２４１、２４２の双方に、それぞれ複数個設置されている。しかしながら、圧力センサ２５は、金型２４１、２４２の一方に１個のみ設置される構成であってもよい。 The pressure sensor 25 is installed in the region of the molds 241 and 242 facing the cavity C1. The pressure sensor 25 detects the pressure in the cavity C1. In particular, the pressure sensor 25 detects the pressure of the molding material (melted state or solidified state, or a state in which the molten state and the solidified state are mixed) supplied into the cavity C1. The pressure sensor 25 outputs a detection signal related to pressure to the control unit 271. In this embodiment, a plurality of pressure sensors 25 are installed on both the molds 241 and 242, respectively. However, only one pressure sensor 25 may be installed in one of the molds 241 and 242.
　温度センサ２６は、金型２４１に内蔵されており、金型２４１の温度を検出する。温度センサ２６は、温度に関する検出信号を制御部２７１に出力する。なお、温度センサ２６は、金型２４１のうちキャビティＣ１に面する領域に設置されてもよいし、金型２４２に設置されてもよい。また、温度センサ２６は、シリンダ２２２内に設置されてもよい。すなわち、温度センサ２６は、キャビティＣ１内に供給される成形材料の温度を直接的に又は間接的に検出することができればよい。 The temperature sensor 26 is built in the mold 241 and detects the temperature of the mold 241. The temperature sensor 26 outputs a detection signal related to temperature to the control unit 271. The temperature sensor 26 may be installed in a region of the mold 241 facing the cavity C1 or may be installed in the mold 242. Further, the temperature sensor 26 may be installed in the cylinder 222. That is, the temperature sensor 26 may be able to directly or indirectly detect the temperature of the molding material supplied into the cavity C1.
　＜成形装置による製造方法＞
　図２から図５（ｂ）を適宜参照しながら、成形装置２０による成型品の製造方法について説明する。成形装置２０による成型品の製造方法は、前工程ＳＴ１と、型締め工程ＳＴ２と、充填工程ＳＴ３と、保圧工程ＳＴ４と、保圧解除工程ＳＴ５と、離型工程ＳＴ６とが、この順で実行される。本実施形態において、成型品は、転がり軸受に用いられる樹脂製の保持器である。しかしながら、これは成型品の一例であり、本開示に係る成形装置により成形される成型品は、その他の形状及び用途の成型品であってもよい。 <Manufacturing method using molding equipment>
A method for manufacturing a molded product by the molding apparatus 20 will be described with reference to FIGS. 2 to 5 (b) as appropriate. As for the method of manufacturing the molded product by the molding apparatus 20, the pre-process ST1, the mold clamping process ST2, the filling process ST3, the pressure holding process ST4, the pressure holding release process ST5, and the mold release process ST6 are performed in this order. Will be executed. In this embodiment, the molded product is a resin cage used for rolling bearings. However, this is an example of a molded product, and the molded product molded by the molding apparatus according to the present disclosure may be a molded product having another shape and use.
　はじめに、前工程ＳＴ１が実行される。前工程ＳＴ１では、モータ２２６によりスクリュ２２３が回転し、ヒータ２２８によりシリンダ２２２内が加熱されている状態で、ホッパ２２１から成形材料のペレットがシリンダ２２２内へ供給される。成形材料のペレットは、スクリュ２２３の回転に伴う摩擦熱と、ヒータ２２８による加熱とにより、シリンダ２２２内において溶融し、溶融状態の成形材料Ｌ１となる。シリンダ２２２内に、所定量の成形材料Ｌ１が貯留されると、前工程ＳＴ１が終了する。 First, the pre-process ST1 is executed. In the previous step ST1, the screw 223 is rotated by the motor 226, and the 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 228. The pellets of the molding material are melted in the cylinder 222 by the frictional heat accompanying the rotation of the screw 223 and the heating by the heater 228 to become the molding material L1 in the molten state. When a predetermined amount of the molding material L1 is stored in the cylinder 222, the previous step ST1 is completed.
　次に、型締め工程ＳＴ２が開始されると、図２の状態の成形装置２０において、制御部２７１の動作指令によりボールねじ２３４が軸方向他方側に移動し、図３に示すように金型２４１を金型２４２に接触させる。このように金型２４１と金型２４２とを組み合わせた状態で、さらにボールねじ２３４が軸方向他方側へ所定の型締め力により金型２４１を金型２４２へ押さえつける。すなわち、複数の金型２４１、２４２を締め付ける。これにより、複数の金型２４１、２４２の間にキャビティＣ１が形成される。以上により、型締め工程ＳＴ２が終了する。 Next, when the mold clamping step ST2 is started, in the molding apparatus 20 in the state of FIG. 2, the ball screw 234 is moved to the other side in the axial direction by the operation command of the control unit 271, and the mold is as shown in FIG. The 241 is brought into contact with the mold 242. In the state where the mold 241 and the mold 242 are combined in this way, the ball screw 234 further presses the mold 241 against the mold 242 toward the other side in the axial direction by a predetermined mold tightening force. That is, the plurality of molds 241 and 242 are tightened. As a result, the cavity C1 is formed between the plurality of molds 241 and 242. With the above, the mold clamping step ST2 is completed.
　ここで、型締め力は、成形条件のひとつであり、金型２４１、２４２の形状等、その他の成形条件に応じて決定される。型締め力は、型締め力センサ２３６により検出される。 Here, the mold clamping force is one of the molding conditions, and is determined according to other molding conditions such as the shape of the molds 241 and 242. The mold clamping force is detected by the mold clamping force sensor 236.
　続いて、充填工程ＳＴ３が開始されると、上記の型締め力を維持している状態で、ボールねじ２２５が軸方向一方側へ移動する。これにより、スクリュ２２３が軸方向一方側へ成形材料Ｌ１を押し、シリンダ２２２のノズル２２４から金型２４２の流路２４３を介してキャビティＣ１へ成形材料Ｌ１が射出される（充填動作）。 Subsequently, when the filling step ST3 is started, the ball screw 225 moves to one side in the axial direction while maintaining the above-mentioned mold clamping force. As a result, the screw 223 pushes the molding material L1 to one side in the axial direction, and the molding material L1 is ejected from the nozzle 224 of the cylinder 222 to the cavity C1 via the flow path 243 of the mold 242 (filling operation).
　図４は、充填工程ＳＴ３の終了時の様子を模式的に示す金型部２４の断面図である。流路２４３は、ノズル２２４側に開口する第１開口部２４３ａと、キャビティＣ１側に開口する第２開口部２４３ｂとを有する。本実施形態において、キャビティＣ１は環状に形成されている。図４に示すように、流路２４３からキャビティＣ１へ成形材料Ｌ１（より具体的には、溶融状態の成形材料）が供給され、キャビティＣ１内がすべて成形材料Ｌ１により充填されると、充填工程ＳＴ３が終了する。また、充填工程ＳＴ３において、溶融状態の成形材料Ｌ１は、金型２４１、２４２の表面付近から徐々に固化しながらキャビティＣ１内へ供給される。 FIG. 4 is a cross-sectional view of the mold portion 24 schematically showing the state at the end of the filling step ST3. The flow path 243 has a first opening 243a that opens on the nozzle 224 side and a second opening 243b that opens on the cavity C1 side. In this embodiment, the cavity C1 is formed in an annular shape. As shown in FIG. 4, when the molding material L1 (more specifically, the molding material in a molten state) is supplied from the flow path 243 to the cavity C1, and the inside of the cavity C1 is completely filled with the molding material L1, the filling step is performed. ST3 ends. Further, in the filling step ST3, the molten molding material L1 is supplied into the cavity C1 while gradually solidifying from the vicinity of the surface of the molds 241 and 242.
　続いて、保圧工程ＳＴ４が開始されると、スクリュ２２３がさらに軸方向一方側へ成形材料Ｌ１を押し、シリンダ２２２のノズル２２４からキャビティＣ１へ成形材料Ｌ１がさらに射出される。これにより、キャビティＣ１内に充填されている成形材料Ｌ１に所定の圧力Ｐｔ１（例えば、数十～数百ＭＰａ）が印加される。そして、スクリュ２２３はこの状態を所定時間保持することで、所定の圧力を所定時間（例えば、数秒間）だけ成形材料Ｌ１に与え続ける（保圧動作）。スクリュ２２３がキャビティＣ１へ成形材料Ｌ１を押し出す圧力（与圧）は、与圧センサ２２７により検出される。 Subsequently, when the pressure holding step ST4 is started, the screw 223 further pushes the molding material L1 to one side in the axial direction, and the molding material L1 is further injected from the nozzle 224 of the cylinder 222 into the cavity C1. As a result, a predetermined pressure Pt1 (for example, several tens to several hundreds of MPa) is applied to the molding material L1 filled in the cavity C1. Then, by holding this state for a predetermined time, the screw 223 continues to apply a predetermined pressure to the molding material L1 for a predetermined time (for example, several seconds) (pressure holding operation). The pressure (pressurization) at which the screw 223 pushes the molding material L1 into the cavity C1 is detected by the pressurization sensor 227.
　続いて、保圧解除工程ＳＴ５が開始されると、スクリュ２２３は軸方向他方側へ移動し、成形材料Ｌ１の圧力の保持を解除する（保圧解除動作）。保圧解除動作後、所定時間が経過してキャビティＣ１内の成形材料Ｌ１の圧力が所定値以下になると、保圧解除工程ＳＴ５が終了する。その後、離型工程ＳＴ６が開始されると、金型部２４が冷却されることで、キャビティＣ１内の成形材料Ｌ１が完全に固化し、成型品が形成される。そして、ボールねじ２３４が軸方向一方側へ移動し、金型２４１が金型２４２から離れることで、成型品が取り出される。なお、金型部２４の冷却は、保圧解除工程ＳＴ５と同時に開始されてもよい。 Subsequently, when the pressure holding release step ST5 is started, the screw 223 moves to the other side in the axial direction to release the pressure holding of the molding material L1 (pressure holding release operation). When a predetermined time elapses after the holding pressure releasing operation and the pressure of the molding material L1 in the cavity C1 becomes equal to or less than a predetermined value, the holding pressure releasing step ST5 ends. After that, when the mold release step ST6 is started, the mold portion 24 is cooled, so that the molding material L1 in the cavity C1 is completely solidified, and a molded product is formed. Then, the ball screw 234 moves to one side in the axial direction, and the mold 241 separates from the mold 242, so that the molded product is taken out. The cooling of the mold portion 24 may be started at the same time as the pressure holding release step ST5.
　図５（ａ）及び図５（ｂ）は、充填工程ＳＴ３、保圧工程ＳＴ４及び保圧解除工程ＳＴ５において、与圧センサ２２７及び圧力センサ２５により検出される圧力の時系列データと当該圧力の傾きの時系列データを示すグラフの一例である。図５（ａ）及び図５（ｂ）では、成型品がバリの発生なく成形された際に得られるグラフを示している。 5 (a) and 5 (b) show the time-series data of the pressure detected by the pressurization sensor 227 and the pressure sensor 25 in the filling step ST3, the pressure holding step ST4, and the pressure holding release step ST5, and the pressure. This is an example of a graph showing time-series data of inclination. 5 (a) and 5 (b) show graphs obtained when the molded product is molded without the occurrence of burrs.
　図５（ａ）を参照する。図５（ａ）において、縦軸は圧力Ｐであり、横軸は時間ｔである。破線により示すグラフ線Ｆ１は、与圧センサ２２７により検出された圧力の時系列データであり、スクリュ２２３がキャビティＣ１へ成形材料Ｌ１を押し出す圧力が表されている。実線により示すグラフ線Ｆ２は、圧力センサ２５により検出された圧力の時系列データであり、キャビティＣ１内の成形材料Ｌ１（溶融状態及び固化状態の少なくとも一方を含む状態）の圧力が表されている。 Refer to FIG. 5 (a). In FIG. 5A, the vertical axis is the pressure P and the horizontal axis is the time t. The graph line F1 shown by the broken line is the time series data of the pressure detected by the pressurization sensor 227, and represents the pressure at which the screw 223 pushes the molding material L1 into the cavity C1. The graph line F2 shown by the solid line is the time series data of the pressure detected by the pressure sensor 25, and represents the pressure of the molding material L1 (the state including at least one of the molten state and the solidified state) in the cavity C1. ..
　図５（ａ）に示すように、与圧（グラフ線Ｆ１）は、充填工程ＳＴ３において０から圧力Ｐｔ１まで上昇し、保圧工程ＳＴ４において圧力Ｐｔ１に所定時間保持され、保圧解除工程ＳＴ５において圧力Ｐｔ１から低下する。保圧解除工程ＳＴ５は、時点Ｘ１から開始される。すなわち、保圧解除動作は、時点Ｘ１に実行される。キャビティＣ１内の成形材料Ｌ１の圧力（グラフ線Ｆ２）は、時点Ｘ１の後、単調減少する。 As shown in FIG. 5A, the pressurization (graph line F1) rises from 0 to the pressure Pt1 in the filling step ST3, is held at the pressure Pt1 for a predetermined time in the pressure holding step ST4, and is held in the pressure Pt1 in the holding pressure releasing step ST5. The pressure drops from Pt1. The holding pressure release step ST5 is started from the time point X1. That is, the holding pressure release operation is executed at the time point X1. The pressure of the molding material L1 in the cavity C1 (graph line F2) decreases monotonically after the time point X1.
　図５（ｂ）を参照する。図５（ｂ）において、縦軸は圧力の傾き（ｄＰ／ｄｔ）であり、横軸は時間ｔである。図５（ｂ）は、時点Ｘ１の後におけるグラフ線Ｆ２を時間微分することで取得される圧力の傾きのグラフ線Ｆ２ａを示している。すなわち、グラフ線Ｆ２ａは、キャビティＣ１内の成形材料Ｌ１の圧力の変化を表すグラフである。時点Ｘ１の後、グラフ線Ｆ２は単調減少するため、グラフ線Ｆ２ａは０より小さい値となる。 Refer to FIG. 5 (b). In FIG. 5B, the vertical axis is the slope of pressure (dP / dt), and the horizontal axis is time t. FIG. 5B shows the graph line F2a of the slope of the pressure obtained by time-differentiating the graph line F2 after the time point X1. That is, the graph line F2a is a graph showing the change in the pressure of the molding material L1 in the cavity C1. After the time point X1, the graph line F2 decreases monotonically, so that the graph line F2a becomes a value smaller than 0.
　＜バリの発生時の圧力変化の説明＞
　成型品にバリが発生する場合、キャビティＣ１内の成形材料Ｌ１の圧力の変化は、図５（ｂ）のグラフ線Ｆ２ａとは異なる挙動を示す。図６から図８（ｂ）を適宜参照しながら、成型品にバリが発生する場合のキャビティＣ１内の成形材料Ｌ１の圧力変化について説明する。 <Explanation of pressure change when burrs occur>
When burrs are generated in the molded product, the change in the pressure of the molding material L1 in the cavity C1 behaves differently from the graph line F2a in FIG. 5 (b). The pressure change of the molding material L1 in the cavity C1 when burrs are generated in the molded product will be described with reference to FIGS. 6 to 8 (b) as appropriate.
　図６は、成型品にバリが発生する場合の、充填工程ＳＴ３の終了時の様子を模式的に示す金型部２４の断面図である。図６の例では、下部に拡大して示すように、金型２４１と金型２４２の間に異物Ｆｍ１が挟まっている。このため、金型２４１と金型２４２の間には、キャビティＣ１の他に、意図しない隙間Ｃ２が形成されてしまう。また、キャビティＣ１の体積も、隙間Ｃ２が形成される幅Ｗ１分だけ図４の例よりも増えてしまう。そして、充填工程ＳＴ３においてキャビティＣ１及び隙間Ｃ２に成形材料Ｌ１が充填される。 FIG. 6 is a cross-sectional view of the mold portion 24 schematically showing the state at the end of the filling step ST3 when burrs are generated in the molded product. In the example of FIG. 6, the foreign matter Fm1 is sandwiched between the molds 241 and the molds 242, as shown in an enlarged manner at the bottom. Therefore, an unintended gap C2 is formed between the mold 241 and the mold 242 in addition to the cavity C1. Further, the volume of the cavity C1 is also increased by the width W1 at which the gap C2 is formed as compared with the example of FIG. Then, in the filling step ST3, the molding material L1 is filled in the cavity C1 and the gap C2.
　図７は、成型品にバリが発生する場合の、保圧解除工程ＳＴ５の様子を模式的に示す断面図である。図７は、図６の拡大図と同じ領域を示している。保圧工程ＳＴ４において、成形材料Ｌ１に所定の圧力Ｐｔ１が保持されているとき、成形材料Ｌ１は金型２４１、２４２のキャビティＣ１及び隙間Ｃ２に露出している面（例えば、金型２４１の露出面２４１ａ）を押している。特に、バリが発生しない場合よりも隙間Ｃ２がある分だけ、成形材料Ｌ１が露出面２４１ａを押す力は大きくなっている。 FIG. 7 is a cross-sectional view schematically showing the state of the pressure holding release step ST5 when burrs are generated in the molded product. FIG. 7 shows the same area as the enlarged view of FIG. In the pressure holding step ST4, when the predetermined pressure Pt1 is held by the molding material L1, the molding material L1 is exposed to the surfaces (for example, the mold 241) exposed to the cavities C1 and the gaps C2 of the molds 241 and 242. The surface 241a) is pressed. In particular, the force with which the molding material L1 pushes the exposed surface 241a is larger by the amount of the gap C2 than when burrs are not generated.
　そして、保圧解除工程ＳＴ５により成形材料Ｌ１の圧力Ｐｔ１の保持が解除されると、成形材料Ｌ１が露出面２４１ａ等を押す力が弱まり、型締め部２３の型締め力が当該押す力よりも強くなることで、図７の矢印ＡＲ１に示す方向に露出面２４１ａが移動する。図７では、保圧工程ＳＴ４時の露出面２４１ａの位置を二点鎖線により示し、保圧解除工程ＳＴ５後の露出面２４１ａの位置を実線により示している。すなわち、保圧解除動作により、金型２４１がわずかに締まり、キャビティＣ１の体積が減少する。また、隙間Ｃ２の体積も減少する。これにより、保圧解除動作の直後、露出面２４１ａ付近の領域Ｃ１ａにおける成形材料Ｌ１の密度は、保圧解除動作前よりも高くなる。この結果、圧力センサ２５により検出される成形材料Ｌ１の圧力は、領域Ｃ１ａにおける成形材料Ｌ１の密度が高くなった分だけ高くなる。 When the holding of the pressure Pt1 of the molding material L1 is released by the pressure holding release step ST5, the force with which the molding material L1 pushes the exposed surface 241a or the like weakens, and the mold clamping force of the mold clamping portion 23 becomes larger than the pressing force. By becoming stronger, the exposed surface 241a moves in the direction indicated by the arrow AR1 in FIG. In FIG. 7, the position of the exposed surface 241a during the pressure holding step ST4 is indicated by a two-dot chain line, and the position of the exposed surface 241a after the pressure holding release step ST5 is indicated by a solid line. That is, the pressure holding release operation slightly tightens the mold 241 and reduces the volume of the cavity C1. In addition, the volume of the gap C2 is also reduced. As a result, immediately after the pressure holding release operation, the density of the molding material L1 in the region C1a near the exposed surface 241a becomes higher than that before the pressure holding release operation. As a result, the pressure of the molding material L1 detected by the pressure sensor 25 increases as the density of the molding material L1 in the region C1a increases.
　図８（ａ）及び図８（ｂ）は、成型品にバリが発生する場合に、充填工程ＳＴ３、保圧工程ＳＴ４及び保圧解除工程ＳＴ５において、与圧センサ２２７及び圧力センサ２５により検出される圧力の時系列データと当該圧力の傾きの時系列データを示すグラフの一例である。グラフの縦軸及び横軸は、図５（ａ）及び図５（ｂ）の例と同様である。また、与圧センサ２２７により検出された圧力の時系列データであるグラフ線Ｆ１は、図５（ａ）の例と同様である。 8 (a) and 8 (b) are detected by the pressurization sensor 227 and the pressure sensor 25 in the filling step ST3, the pressure holding step ST4, and the pressure releasing step ST5 when burrs are generated in the molded product. This is an example of a graph showing the time-series data of the pressure and the time-series data of the slope of the pressure. The vertical and horizontal axes of the graph are the same as those in FIGS. 5 (a) and 5 (b). Further, the graph line F1 which is the time series data of the pressure detected by the pressurization sensor 227 is the same as the example of FIG. 5A.
　図８（ａ）において、実線により示すグラフ線Ｆ３は、圧力センサ２５により検出された圧力の時系列データであり、キャビティＣ１内の成形材料Ｌ１の圧力が表されている。図８（ａ）中の矢印ＡＲ２により示すように、圧力センサ２５により検出される成形材料Ｌ１の圧力は、保圧解除動作後（時点Ｘ１後）に一瞬だけ上昇している。すなわち、バリが発生する場合、保圧解除動作後のグラフ線Ｆ３は、非単調減少の傾向を示す。 In FIG. 8A, the graph line F3 shown by the solid line is the time series data of the pressure detected by the pressure sensor 25, and represents the pressure of the molding material L1 in the cavity C1. As shown by the arrow AR2 in FIG. 8A, the pressure of the molding material L1 detected by the pressure sensor 25 rises only momentarily after the pressure holding release operation (after the time point X1). That is, when burrs occur, the graph line F3 after the holding pressure release operation shows a tendency of non-monotonic decrease.
　図８（ｂ）を参照する。図８（ｂ）は、時点Ｘ１の後におけるグラフ線Ｆ３を時間微分することで取得される圧力の傾きのグラフ線Ｆ３ａを示している。すなわち、グラフ線Ｆ３ａは、キャビティＣ１内の成形材料Ｌ１の圧力の変化を表すグラフである。時点Ｘ１の後、グラフ線Ｆ３は一瞬だけ増加した後に単調減少するため、グラフ線Ｆ３ａには０より大きい値を有する領域がある。 Refer to FIG. 8 (b). FIG. 8B shows the graph line F3a of the slope of the pressure obtained by time-differentiating the graph line F3 after the time point X1. That is, the graph line F3a is a graph showing the change in the pressure of the molding material L1 in the cavity C1. Since the graph line F3 increases for a moment and then decreases monotonically after the time point X1, the graph line F3a has a region having a value larger than 0.
　なお、図７に示す挙動が生じる場合であっても、保圧解除動作後のキャビティＣ１内の成形材料Ｌ１の圧力の減少の程度の方が、キャビティＣ１における成形材料Ｌ１の密度の上昇の程度よりも大きい場合、図８（ａ）の矢印ＡＲ２に示す圧力の増加が生じない場合がある。このような場合であっても、保圧解除動作後において、バリが発生する場合の圧力の時系列データには、領域Ｃ１ａにおける成形材料Ｌ１の密度の上昇がある分だけ、図５（ａ）に示す圧力の時系列データと比べて、一瞬だけ圧力の減少が緩やかになる部分が生じる。すなわち、保圧解除動作後において、バリが発生する場合の圧力の傾きは、バリが発生しない場合の圧力の傾きと比べて、大きい値となる。 Even when the behavior shown in FIG. 7 occurs, the degree of decrease in the pressure of the molding material L1 in the cavity C1 after the pressure holding release operation is the degree of increase in the density of the molding material L1 in the cavity C1. If it is larger than, the pressure increase shown by the arrow AR2 in FIG. 8A may not occur. Even in such a case, the time-series data of the pressure when burrs occur after the pressure holding release operation includes an increase in the density of the molding material L1 in the region C1a, and FIG. Compared with the time-series data of the pressure shown in, there is a part where the pressure decrease becomes slow for a moment. That is, after the pressure holding release operation, the slope of the pressure when burrs are generated becomes a larger value than the slope of the pressure when burrs are not generated.
　上記に説明したように、発明者らは、鋭意研究の結果、保圧解除動作後のキャビティＣ１内の成形材料Ｌ１の圧力が、正常時（バリ非発生時）には単調減少する一方で、バリ発生時には成形材料Ｌ１の圧力が一瞬だけ増加、非減少、又は減少の程度が小さくなる（圧力の傾きが一瞬だけ大きくなる）ことを発見した。すなわち、保圧解除動作後の成形材料Ｌ１の圧力の変化に着目すれば、バリの状態を予測できることを発見した。 As explained above, as a result of diligent research, the inventors have found that the pressure of the molding material L1 in the cavity C1 after the pressure holding release operation decreases monotonically under normal conditions (when burrs do not occur), while it becomes monotonous. It was discovered that when burrs occur, the pressure of the molding material L1 increases, does not decrease, or decreases for a moment (the pressure gradient increases for a moment). That is, it was discovered that the state of burrs can be predicted by paying attention to the change in the pressure of the molding material L1 after the pressure holding release operation.
　そこで、本実施形態に係る成形システム１０では、学習装置３０において保圧解除動作後のキャビティＣ１内の成形材料Ｌ１の圧力の変化とバリの状態との相関関係を学習させた学習済みモデルＴｍ１を生成し、異常予測装置４０において学習済みモデルＴｍ１と保圧解除動作後の成形材料Ｌ１の圧力の変化とに基づいて、バリの状態を予測するための予測情報を取得する。以下、学習装置３０及び異常予測装置４０について説明する。 Therefore, in the molding system 10 according to the present embodiment, the learned model Tm1 is trained in the learning device 30 to learn the correlation between the pressure change of the molding material L1 in the cavity C1 after the pressure holding release operation and the burr state. The generation and the abnormality prediction device 40 acquire the prediction information for predicting the state of burrs based on the trained model Tm1 and the change in the pressure of the molding material L1 after the pressure holding release operation. Hereinafter, the learning device 30 and the abnormality prediction device 40 will be described.
　＜学習装置の説明＞
　図９は、本実施形態に係る学習装置３０の機能構成を示すブロック図である。学習装置３０は、訓練データ取得部３１と、学習演算部３２と、成形情報記憶部３３と、学習済みモデル記憶部３４とを有する。これらの各部は、ＣＰＵ等の演算部とＨＤＤ等の記憶部とを有するコンピュータ装置により実現される。 <Explanation of learning device>
FIG. 9 is a block diagram showing a functional configuration of the learning device 30 according to the present embodiment. The learning device 30 includes 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 an HDD.
　成形情報記憶部３３には、各種の成形情報が記憶されている。成形情報は、例えば各種の第１情報と第２情報とを対応付けしたテーブル形式の情報である。例えば、第１情報が金型の種類である場合、第２情報には金型の各種寸法、キャビティＣ１の体積が含まれる。第１情報が成形材料の種類又はロット番号である場合、第２情報には成形材料の物性（粘度、含有水分等）が含まれる。 Various molding information is stored in the molding information storage unit 33. The molding information is, for example, table-type information in which various types of first information and second information are associated with each other. For example, when the first information is the type of the 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 (viscosity, moisture content, etc.) of the molding material.
　訓練データ取得部３１は、成形システム１０の各部から訓練データに関する情報を取得する。訓練データは、後述の評価値と、第２評価値と、第３評価値と、成形情報と、バリ情報とを含む。また、訓練データは、型締め力センサ２３６及び与圧センサ２２７においてそれぞれ検出された圧力（型締め力及び与圧）を含む。 The training data acquisition unit 31 acquires information on the training data from each unit of the molding system 10. The training data includes an evaluation value described later, a second evaluation value, a third evaluation value, molding information, and burr information. The training data also includes the pressures (molding force and pressurization) detected by the mold clamping force sensor 236 and the pressurization sensor 227, respectively.
　例えば、訓練データ取得部３１は、圧力センサ２５及び与圧センサ２２７においてそれぞれ検出された圧力に基づいて、保圧解除動作後の圧力の変化に関する評価値を取得する。また、訓練データ取得部３１は、温度センサ２６において検出された温度に関する第２評価値を取得する。また、訓練データ取得部３１は、図示省略するその他のセンサ（例えば、湿度センサ）において検出された成形装置２０の周辺及び内部の環境に関する第３評価値を取得する。 For example, the training data acquisition unit 31 acquires an evaluation value regarding a change in pressure after the pressurization release operation based on the pressure detected by the pressure sensor 25 and the pressurization sensor 227, respectively. In addition, the training data acquisition unit 31 acquires a second evaluation value regarding the temperature detected by the temperature sensor 26. In addition, the training data acquisition unit 31 acquires a third evaluation value regarding the surrounding and internal environment of the molding apparatus 20 detected by another sensor (for example, a humidity sensor) (not shown).
　図１０（ａ）及び図１０（ｂ）は、評価値について説明する説明図である。図１０（ａ）は、圧力の傾きの時系列データを示すグラフである。図１０（ａ）の縦軸は圧力の傾き（ｄＰ／ｄｔ）であり、横軸は時間ｔである。成型品を成形装置２０により製造した際に、圧力センサ２５により圧力の時系列データが取得される。そして、当該圧力の時系列データに基づいて、圧力の傾きの時系列データが取得される。 10 (a) and 10 (b) are explanatory views for explaining the evaluation values. FIG. 10A is a graph showing time-series data of the slope of pressure. The vertical axis of FIG. 10A is the slope of pressure (dP / dt), and the horizontal axis is time t. When the molded product is manufactured by the molding apparatus 20, the pressure sensor 25 acquires the time series data of the pressure. Then, based on the time-series data of the pressure, the time-series data of the slope of the pressure is acquired.
　図４に示すように、圧力センサ２５は複数設けられているため、成型品を１個製造すると、圧力の時系列データは複数取得される。本実施形態では、複数の圧力センサ２５によりそれぞれ取得される複数の圧力の時系列データから、それぞれ複数の圧力の傾きの時系列データを取得し、それぞれ複数の評価値を取得する。以下の説明では、複数の圧力センサ２５のうち、１個の圧力センサ２５により得られる圧力の時系列データに着目して説明する。なお、複数の圧力センサ２５によりそれぞれ取得される複数の圧力の時系列データについて平均値を求めた後、平均圧力の傾きの時系列データと、１個の評価値を取得するように構成してもよい。 As shown in FIG. 4, since a plurality of pressure sensors 25 are provided, when one molded product is manufactured, a plurality of pressure time series data are acquired. In the present embodiment, from the time-series data of the plurality of pressures acquired by the plurality of pressure sensors 25, the time-series data of the inclinations of the plurality of pressures are acquired, and a plurality of evaluation values are acquired respectively. In the following description, the time series data of the pressure obtained by one pressure sensor 25 among the plurality of pressure sensors 25 will be focused on. In addition, after obtaining the average value for the time series data of a plurality of pressures acquired by the plurality of pressure sensors 25, the time series data of the inclination of the average pressure and one evaluation value are acquired. May be good.
　図１０（ａ）には、３個の成型品を成形した際に取得される圧力の傾きの時系列データの一例を示している。グラフ線Ｆ３１（実線）、グラフ線Ｆ３２（二点鎖線）及びグラフ線Ｆ３３（破線）は、いずれも成型品にバリが発生する場合の圧力の傾きに関するグラフであり、保圧解除動作後（時点Ｘ１後）の圧力の傾きのうち、傾きが０以上になる領域を拡大して示すグラフである。ここで、時点ｔにおける圧力の傾き（ｄＰ／ｄｔ）は、例えば、圧力の時系列データにおいて、時点ｔから時点（ｔ＋ｄｔ）までの圧力の変化量ｄＰを、時間の変化量ｄｔで除算することにより求められる。 FIG. 10A shows an example of time-series data of the slope of the pressure acquired when three molded products are molded. The graph line F31 (solid line), the graph line F32 (two-dot chain line), and the graph line F33 (broken line) are all graphs relating to the slope of the pressure when burrs occur in the molded product, and are after the pressure holding release operation (time point). It is a graph which enlarges and shows the region where the slope becomes 0 or more in the slope of pressure (after X1). Here, the pressure gradient (dP / dt) at the time point t is, for example, the amount of change in pressure dP from the time point t to the time point (t + dt) in the time series data of pressure divided by the amount of change in time dt. Demanded by.
　評価値は、例えば圧力の傾きの最大値である。ある成型品を成形し、グラフ線Ｆ３１が取得された場合、評価値はｄＰ１となる。また、他のある成型品を成形し、グラフ線Ｆ３２又はＦ３３が取得された場合、評価値はｄＰ２又はｄＰ３となる。評価値は、訓練データ取得部３１において、圧力の時系列データに基づいて演算することで取得される。 The evaluation value is, for example, the maximum value of the pressure gradient. When a certain molded product is molded and the graph line F31 is acquired, the evaluation value is dP1. Further, when a certain other molded product is molded and the graph line F32 or F33 is acquired, the evaluation value becomes dP2 or dP3. The evaluation value is acquired by the training data acquisition unit 31 by calculating based on the pressure time series data.
　なお、評価値は、圧力の傾き（ｄＰ／ｄｔ）に限られず、圧力の変化量ｄＰであってもよい。また、評価値は、図１０（ａ）に示すように、グラフ線Ｆ３１、Ｆ３２、Ｆ３３のそれぞれの半値全幅Ｗ１、Ｗ２、Ｗ３であってもよい。半値全幅Ｗ１は、グラフ線Ｆ３１のうち圧力の傾きが最大値ｄＰ１の半分（ｄＰ１／２）となる時点の幅である。すなわち、評価値は、圧力の傾きが０以上となる領域の幅であってもよい。 The evaluation value is not limited to the slope of pressure (dP / dt), and may be the amount of change in pressure dP. Further, as shown in FIG. 10A, the evaluation value may be the full width at half maximum W1, W2, W3 of the graph lines F31, F32, and F33, respectively. The full width at half maximum W1 is the width of the graph line F31 at the time when the slope of the pressure becomes half of the maximum value dP1 (dP1 / 2). That is, the evaluation value may be the width of the region where the slope of the pressure is 0 or more.
　また、図１０（ｂ）に示すように、評価値は、圧力の時系列データの所定の２点間の圧力差であってもよいし、所定の２点間の圧力の変化率であってもよい。図１０（ｂ）は、図８（ａ）のグラフ線Ｆ３のうち、矢印ＡＲ２により示す領域を拡大して示すグラフである。図１０（ｂ）に示すように、成型品にバリが発生する場合、保圧解除動作（時点Ｘ１）後のグラフ線Ｆ３には極小（時点Ｘａ、圧力Ｐ１）と、極大（時点Ｘｂ、圧力Ｐ２）が現れる。評価値は、極小における圧力Ｐ１と極大における圧力Ｐ２との差（Ｐ２－Ｐ１）であってもよい。また、評価値は、当該差（Ｐ２－Ｐ１）を極小と極大の時間差（Ｘｂ－Ｘａ）により除算することで得られる圧力の変化率（（Ｐ２－Ｐ１）／（Ｘｂ－Ｘａ））であってもよい。すなわち、評価値は保圧解除動作後における圧力の変化を表す値であればよい。 Further, as shown in FIG. 10B, the evaluation value may be a pressure difference between two predetermined points in the time series data of pressure, or a rate of change in pressure between the predetermined two points. May be good. FIG. 10B is a graph showing an enlarged area indicated by the arrow AR2 in the graph line F3 of FIG. 8A. As shown in FIG. 10B, when burrs occur in the molded product, the graph line F3 after the pressure holding release operation (time point X1) has a minimum (time point Xa, pressure P1) and a maximum (time point Xb, pressure). P2) appears. The evaluation value may be the difference (P2-P1) between the pressure P1 at the minimum and the pressure P2 at the maximum. The evaluation value is the rate of change in pressure ((P2-P1) / (Xb-Xa)) obtained by dividing the difference (P2-P1) by the time difference (Xb-Xa) between the minimum and the maximum. You may. That is, the evaluation value may be a value indicating a change in pressure after the pressure holding release operation.
　図９を参照する。訓練データ取得部３１は、入力部５０にオペレータが入力する情報を取得する。オペレータが入力する情報は、例えば成型品のバリの状態に関するバリ情報や、成形情報記憶部３３に記憶されている第１情報である。訓練データ取得部３１は、入力部５０に入力された第１情報に基づいて、成形情報記憶部３３から当該第１情報に対応する第２情報を取得する。例えば、入力部５０に成形材料のロット番号が入力された場合に、訓練データ取得部３１は成形情報記憶部３３から当該ロット番号に対応する成形材料の物性に関する情報を取得する。 Refer to FIG. The training data acquisition unit 31 acquires information input by the operator to the input unit 50. The information input by the operator is, for example, burr information regarding the burr state of the molded product or first information stored in the molding information storage unit 33. The training data acquisition unit 31 acquires the second information corresponding to the first information from the molding information storage unit 33 based on the first information input to the input unit 50. For example, when a lot number of a molding material is input to the input unit 50, the training data acquisition unit 31 acquires information on the physical properties of the molding material corresponding to the lot number from the molding information storage unit 33.
　バリ情報は、例えばバリの有無、バリの大きさ（例えば、バリの長さ、幅等）を数値化した情報（例えば、ダミー変数）である。バリの大きさは、バリの長さそのものを数値化してもよいし、バリの大きさの程度を数値化してもよい。例えば、バリの大きさの程度を大・中・小・無しの４グループに分け、ダミー変数を「大＝３」、「中＝２」、「小＝１」、「無し＝０」と割り当てることで、数値化してもよい。オペレータは、成形装置２０により成形された成型品の外観を実際に検査することで、バリ情報を取得し、当該バリ情報を入力部５０へ入力する。 The burr information is, for example, information (for example, a dummy variable) that quantifies the presence or absence of burrs and the size of burrs (for example, the length and width of burrs). As for the size of the burr, the length of the burr itself may be quantified, or the degree of the size of the burr may be quantified. For example, the degree of burr size is divided into 4 groups of large, medium, small, and none, and dummy variables are assigned as "large = 3", "medium = 2", "small = 1", and "none = 0". By doing so, it may be quantified. The operator acquires burr information by actually inspecting the appearance of the molded product molded by the molding apparatus 20, and inputs the burr information to the input unit 50.
　学習演算部３２は、訓練データに基づいて、教師あり機械学習を行う演算をすることで、評価値と成型品のバリの状態との相関関係をモデル化した学習済みモデルを生成する。本実施形態では、機械学習モデルとして、畳み込みニューラルネットワーク（ＣＣＮ：Convolutional Neural Network）を用いるが、その他のモデルを用いてもよい。例えば、データのグループ分けに関するモデルである回帰木モデルであってもよい。 The learning calculation unit 32 generates a learned model that models the correlation between the evaluation value and the burr state of the molded product by performing a supervised machine learning calculation based on the training data. In this embodiment, a convolutional neural network (CCN) is used as the machine learning model, but other models may be used. For example, it may be a regression tree model that is a model for grouping data.
　具体的には、評価情報を説明変数とし、成型品のバリの状態に関するバリ情報を目的関数とすることで、評価情報と成型品のバリの状態との相関関係をモデル化する。上記で説明したとおり、評価情報は、評価値（保圧解除動作後の圧力の変化に関する値）、第２評価値（温度に関する値）、第３評価値（湿度等に関する値）、成形情報（成形材料の粘度等）、型締め力及び与圧を含む。 Specifically, by using the evaluation information as an explanatory variable and the burr information regarding the burr state of the molded product as the objective function, the correlation between the evaluation information and the burr state of the molded product is modeled. As explained above, the evaluation information includes the evaluation value (value related to the change in pressure after the holding pressure release operation), the second evaluation value (value related to temperature), the third evaluation value (value related to humidity, etc.), and molding information (value related to humidity, etc.). Includes molding material viscosity, etc.), mold clamping force and pressurization.
　学習演算部３２により生成された学習済みモデルＴｍ１は、学習済みモデル記憶部３４に記憶される。学習済みモデル記憶部３４に記憶された学習済みモデルＴｍ１は、学習装置３０に新たな情報が入力され、訓練データ取得部３１において新たな訓練データが取得されると、当該訓練データの内容に応じて適宜更新される。また、学習済みモデルＴｍ１は、学習装置３０から後述の異常予測装置４０へ送信され、異常予測装置４０の学習済みモデル記憶部４５にも記憶される。本実施形態の学習済みモデルＴｍ１は、例えば非一時的なコンピュータ可読媒体等の、任意の記憶媒体に格納されて提供されてもよい。 The learned model Tm1 generated by the learning calculation unit 32 is stored in the learned 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 responds to the content of the training data. Will be updated as appropriate. Further, the learned model Tm1 is transmitted from the learning device 30 to the abnormality prediction device 40 described later, and is also stored in the learned model storage unit 45 of the abnormality prediction device 40. The trained model Tm1 of the present embodiment may be stored and provided in an arbitrary storage medium such as a non-temporary computer-readable medium.
　＜学習済みモデルの生成方法＞
　次に、学習装置３０による学習済みモデルＴｍ１の生成方法について説明する。学習済みモデルＴｍ１の生成方法は、訓練データ取得工程と、学習演算工程とを備える。はじめに、訓練データ取得工程が開始されると、訓練データ取得部３１は、訓練データを取得する。例えば、キャビティＣ１内の成形材料Ｌ１の圧力を検出する圧力センサ２５により検出された圧力の時系列データに基づいて、保圧解除工程後の成形材料Ｌ１の圧力の変化に関する評価値を取得する。また、当該評価値が取得された成型品をオペレータが実際に検査することでバリ情報を取得する。次に、学習演算工程が開始されると、学習演算部３２は、訓練データに基づいて学習済みモデルＴｍ１を生成する。 <How to generate a trained model>
Next, a method of generating the trained model Tm1 by the learning device 30 will be described. The method of generating the trained model Tm1 includes a training data acquisition step and a learning calculation step. First, when the training data acquisition process is started, the training data acquisition unit 31 acquires the training data. For example, based on the time series data of the pressure detected by the pressure sensor 25 that detects the pressure of the molding material L1 in the cavity C1, the evaluation value regarding the change in the pressure of the molding material L1 after the pressure holding release step is acquired. In addition, the operator actually inspects the molded product for which the evaluation value has been acquired to acquire burr information. Next, when the learning calculation process is started, the learning calculation unit 32 generates the trained model Tm1 based on the training data.
　＜異常予測装置の説明＞
　図１１は、本実施形態に係る異常予測装置４０の機能構成を示すブロック図である。異常予測装置４０は、データ取得部４１と、異常予測部４２と、出力部４３と、成形情報記憶部４４と、学習済みモデル記憶部４５とを有する。これらの各部は、ＣＰＵ等の演算部とＨＤＤ等の記憶部とを有するコンピュータ装置により実現される。演算部は、記憶部に記憶されているプログラムに基づいて、後述のデータ取得処理と、異常予測処理とを実行する。本実施形態のプログラムは、例えば非一時的なコンピュータ可読媒体等の、任意の記憶媒体に格納されて提供されてもよい。 <Explanation of abnormality prediction device>
FIG. 11 is a block diagram showing a functional configuration of the abnormality prediction device 40 according to the present embodiment. The abnormality prediction device 40 includes a data acquisition unit 41, an abnormality prediction unit 42, an output unit 43, a molding information storage unit 44, and a learned 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 an HDD. The calculation unit executes the data acquisition process and the abnormality prediction process, which will be described later, based on the program stored in the storage unit. The program of the present embodiment may be stored and provided on an arbitrary storage medium such as a non-transitory computer-readable medium.
　成形情報記憶部４４には、成形情報記憶部３３と同様に、各種の第１情報と第２情報とを対応付けしたテーブル形式の成形情報が記憶されている。学習済みモデル記憶部４５には、学習装置３０により生成された学習済みモデルＴｍ１が記憶されている。 Similar to the molding information storage unit 33, the molding information storage unit 44 stores table-type molding information in which various first information and the second information are associated with each other. The learned model Tm1 generated by the learning device 30 is stored in the learned model storage unit 45.
　成形情報記憶部４４及び学習済みモデル記憶部４５は、コンピュータ装置のうち、学習装置３０の成形情報記憶部３３及び学習済みモデル記憶部３４と同じ記憶領域により実現されてもよいし、別の記憶領域により実現されてもよい。すなわち、学習装置３０及び異常予測装置４０が、同じ成形情報記憶部３３及び学習済みモデル記憶部３４を共有するように構成されてもよいし、学習装置３０及び異常予測装置４０がそれぞれ独立した成形情報記憶部３３、４４及び学習済みモデル記憶部３４、４５を有するように構成されてもよい。 The molding information storage unit 44 and the learned model storage unit 45 may be realized by the same storage area as the molding information storage unit 33 and the learned model storage unit 34 of the learning device 30 among the computer devices, or may be different storage. It may be realized by the area. That is, the learning device 30 and the abnormality prediction device 40 may be configured to share the same molding information storage unit 33 and the learned model storage unit 34, or the learning device 30 and the abnormality prediction device 40 may be independently molded. It may be configured to have information storage units 33, 44 and trained model storage units 34, 45.
　データ取得部４１は、成形システム１０の各部から異常予測を行うための評価情報を取得するデータ取得処理を実行する。ここで、評価情報は、上記の学習装置３０において取得する評価情報と同様である。すなわち、評価情報は、評価値（保圧解除動作後の圧力の変化に関する値）、第２評価値（温度に関する値）、第３評価値（湿度等に関する値）、成形情報（成形材料の粘度等）、型締め力及び与圧を含む。 The data acquisition unit 41 executes a data acquisition process for acquiring evaluation information for performing abnormality prediction from each unit of the molding system 10. Here, the evaluation information is the same as the evaluation information acquired by the learning device 30 described above. That is, the evaluation information includes the evaluation value (value related to the change in pressure after the pressurization release operation), the second evaluation value (value related to temperature), the third evaluation value (value related to humidity, etc.), and the molding information (viscosity of the molding material). Etc.), including mold clamping force and pressurization.
　例えば、データ取得部４１は、圧力センサ２５及び与圧センサ２２７においてそれぞれ検出された圧力に基づいて、保圧解除動作後の圧力の変化に関する評価値を取得する。評価値は、以上に説明した学習装置３０の訓練データ取得部３１において取得される評価値と同様の値である。すなわち、評価値は、保圧解除動作後の成形材料Ｌ１の圧力の変化に関する値である。 For example, the data acquisition unit 41 acquires an evaluation value regarding a change in pressure after the pressure holding release operation based on the pressure detected by the pressure sensor 25 and the pressurization sensor 227, respectively. The evaluation value is the same value as the evaluation value acquired by the training data acquisition unit 31 of the learning device 30 described above. That is, the evaluation value is a value related to the change in pressure of the molding material L1 after the pressure holding release operation.
　異常予測部４２は、学習済みモデルＴｍ１へ、データ取得部４１により取得された評価情報を入力することで、成型品のバリの状態を予測するための予測情報を取得する異常予測処理を実行する。異常予測処理では、学習済みモデルＴｍ１を生成する際に用いた目的変数に対応する情報が、予測情報として出力される。 The abnormality prediction unit 42 executes an abnormality prediction process for acquiring prediction information for predicting the burr state of the molded product by inputting the evaluation information acquired by the data acquisition unit 41 into the trained model Tm1. .. In the abnormality prediction process, the information corresponding to the objective variable used when the trained model Tm1 is generated is output as the prediction information.
　例えば、目的変数が成型品のバリの大きさである場合、予測情報は、成型品のバリの大きさについて予測される確率を含む情報となる。より具体的には、予測情報は、例えば成型品のバリの長さについて、１ｍｍ以下である第１確率と、１ｍｍより長く５ｍｍ以下である第２確率と、５ｍｍより長い第３確率とを含む。一例として、学習済みモデルＴｍ１へある評価情報を入力すると、学習済みモデルＴｍ１は第１確率が１０％、第２確率が１０％、第３確率が８０％である予測情報を出力する。 For example, when the objective variable is the size of the burr of the molded product, the prediction information is information including the probability of being predicted about the size of the burr of the molded product. More specifically, the prediction information includes, for example, the first probability that the burr length of the molded product is 1 mm or less, the second probability that it is longer than 1 mm and 5 mm or less, and the third probability that it is longer than 5 mm. .. As an example, when some evaluation information is input to the trained model Tm1, the trained model Tm1 outputs prediction information having a first probability of 10%, a second probability of 10%, and a third probability of 80%.
　なお、予測情報は、バリの有無によって分類される確率の情報であってもよい。すなわち、目的変数が成型品のバリの有無である場合、予測情報は、バリがない第４確率と、バリがある第５確率となる。 Note that the prediction information may be information on the probability of being classified according to the presence or absence of burrs. That is, when the objective variable is the presence or absence of burrs on the molded product, the prediction information is the fourth probability without burrs and the fifth probability with burrs.
　出力部４３は、異常予測部４２において取得された予測情報に基づいて、成型品の良否を判定する。例えば、出力部４３は、所定のしきい値と、上記の予測情報に基づいて、良否を判定する。所定のしきい値は、例えば許容されるバリの最大長さである。例えば、許容されるバリの最大長さが５ｍｍである場合、上記の第１確率～第３確率のうち、第３確率が最も高いときに、成型品が不良（バリ不良あり）であると判定する。 The output unit 43 determines the quality of the molded product based on the prediction information acquired by the abnormality prediction unit 42. For example, the output unit 43 determines the quality based on a predetermined threshold value and the above prediction information. The predetermined threshold is, for example, the maximum length of burrs allowed. For example, when the maximum allowable length of burrs is 5 mm, it is determined that the molded product is defective (with burrs) when the third probability is the highest among the first to third probabilities described above. do.
　出力部４３は、成型品の良否に関する判定情報と、予測情報とを表示部６０及び制御部２７１に出力する。表示部６０には、判定情報と、予測情報とが表示される。特に、成型品が不良であると判定されている場合には、表示部６０のディスプレイにおいて赤などの強調色により判定情報を表示し、スピーカにおいてアラートを発報するように構成してもよい。 The output unit 43 outputs the determination information regarding the quality of the molded product and the prediction information to the display unit 60 and the control unit 271. Judgment information and prediction information are displayed on the display unit 60. In particular, when it is determined that the molded product is defective, the determination information may be displayed in a highlighted color such as red on the display of the display unit 60, and an alert may be issued on the speaker.
　また、成型品が不良であると判定されている場合、制御部２７１の動作指令により、不良判定された成型品を成形した成形装置２０を、金型部２４が開放した状態で停止させるように構成してもよい。この場合、オペレータは表示部６０によるアラート等に基づいて、金型部２４を点検し、必要に応じて異物Ｆｍ１（図６）の除去処理を行う。 Further, when it is determined that the molded product is defective, the operation command of the control unit 271 causes the molding apparatus 20 that has molded the molded product determined to be defective to be stopped in a state where the mold portion 24 is open. It may be configured. In this case, the operator inspects the mold unit 24 based on the alert or the like by the display unit 60, and removes the foreign matter Fm1 (FIG. 6) as necessary.
　なお、本実施形態において、出力部４３を設けずに、異常予測部４２において得られた予測情報をそのまま表示部６０に表示するように構成されてもよい。この場合、表示部６０に表示された予測情報に基づいて、オペレータが成型品の良否や、成形装置２０の状態を判断するようにしてもよい。 Note that, in the present embodiment, the prediction information obtained by the abnormality prediction unit 42 may be displayed as it is on the display unit 60 without providing the output unit 43. In this case, the operator may determine the quality of the molded product and the state of the molding apparatus 20 based on the prediction information displayed on the display unit 60.
　＜異常予測装置による異常予測方法＞
　次に、異常予測装置４０による異常予測方法を説明する。異常予測方法は、データ取得工程と、異常予測工程とを備える。データ取得工程が開始されると、データ取得部４１は、評価情報を取得する。例えば、キャビティＣ１内の成形材料Ｌ１の圧力を検出する圧力センサ２５により検出された圧力の時系列データに基づいて、保圧解除工程後の成形材料Ｌ１の圧力の変化に関する評価値を取得する。次に、異常予測工程が開始されると、異常予測部４２は、評価値を含む評価情報に基づいて、成型品のバリの状態を予測するための予測情報を取得する。 <Abnormality prediction method using anomaly prediction device>
Next, the abnormality prediction method by the abnormality prediction device 40 will be described. The abnormality prediction method includes a data acquisition step and an abnormality prediction step. When the data acquisition process is started, the data acquisition unit 41 acquires the evaluation information. For example, based on the time series data of the pressure detected by the pressure sensor 25 that detects the pressure of the molding material L1 in the cavity C1, the evaluation value regarding the change in the pressure of the molding material L1 after the pressure holding release step is acquired. Next, when the abnormality prediction step is started, the abnormality prediction unit 42 acquires prediction information for predicting the burr state of the molded product based on the evaluation information including the evaluation value.
　＜成形システムの作用・効果＞
　本実施形態に係る成形システム１０は、保圧解除動作後の成形材料Ｌ１の圧力の変化に基づいて、バリの状態を予測するための予測情報を取得する。このような構成により、様々な成形条件ごとに基準となる曲線を用意する必要がなく、予測対象となる１つの圧力情報からバリの状態を予測することが可能となる。これにより、より正確にバリの状態を予測することができる。 <Action / effect of molding system>
The molding system 10 according to the present embodiment acquires prediction information for predicting the state of burrs based on the change in the pressure of the molding material L1 after the pressure holding release operation. With such a configuration, it is not necessary to prepare a reference curve for each of various molding conditions, and it is possible to predict the state of burrs from one pressure information to be predicted. This makes it possible to predict the state of burrs more accurately.
　また、本実施形態に係る成形システム１０は、評価値と成型品のバリの状態との相関関係を機械学習させた学習済みモデルＴｍ１へ評価値を入力することで、予測情報を取得する。ここで、学習済みモデルＴｍ１の説明変数は、評価値を含み、学習済みモデルＴｍ１の目的変数は、成型品のバリの状態（例えば、バリの有無、又はバリの大きさ）である。このように構成することで、成形条件にばらつきがある状態であっても、より正確にバリの状態を予測することができる。 Further, the molding system 10 according to the present embodiment acquires prediction information by inputting the evaluation value into the trained model Tm1 in which the correlation between the evaluation value and the burr state of the molded product is machine-learned. Here, the explanatory variable of the trained model Tm1 includes an evaluation value, and the objective variable of the trained model Tm1 is the state of burrs (for example, the presence or absence of burrs or the size of burrs) of the molded product. With this configuration, the burr state can be predicted more accurately even when the molding conditions vary.
　また、保圧解除動作後の成形材料Ｌ１の圧力の変化は、成形材料Ｌ１の温度にも依存する。成形材料Ｌ１の温度が高いほど、成形材料Ｌ１の粘度が低く、成形材料Ｌ１が流動しやすくなる。このため、成形材料Ｌ１の温度が高いほど、バリ発生時の圧力の変化が短時間のうちに急峻に生じやすくなる。本実施形態に係る成形システム１０は、成形材料Ｌ１の温度に関する第２評価値を説明変数に含むことで、上記の相関関係が組み込まれた学習済みモデルとすることが可能となり、より正確にバリの状態を予測することができる。 Further, the change in the pressure of the molding material L1 after the pressure holding release operation also depends on the temperature of the molding material L1. The higher the temperature of the molding material L1, the lower the viscosity of the molding material L1 and the easier it is for the molding material L1 to flow. Therefore, the higher the temperature of the molding material L1, the steeper the change in pressure at the time of burr generation tends to occur in a short time. By including the second evaluation value regarding the temperature of the molding material L1 in the explanatory variables, the molding system 10 according to the present embodiment can be a trained model in which the above correlation is incorporated, and more accurately burrs. The state of can be predicted.
　＜第２実施形態＞
　以上、第１実施形態に係る成形システムを説明した。しかしながら、本開示の実施に関してはこれに限られず、種々の変形を行うことができる。以下、本開示の第２実施形態に係る成形システム１１について、説明する。なお、以下の説明において、第１実施形態から変更のない部分については同じ符号を付し、説明を省略する。 <Second Embodiment>
The molding system according to the first embodiment has been described above. However, the implementation of the present disclosure is not limited to this, and various modifications can be made. Hereinafter, the molding system 11 according to the second embodiment of the present disclosure will be described. In the following description, the parts that are not changed from the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.
　図１２は、第２実施形態に係る成形システム１１を模式的に示すブロック図である。成形システム１１は、複数の成形装置２０と、異常予測装置４０ａと、入力部５０と、表示部６０とを備える。本実施形態において、成形システム１１の異常予測装置４０ａは、評価値と所定の基準値とに基づいて、成型品のバリの状態を予測するための予測情報を取得する。すなわち、成形システム１１は、学習済みモデルＴｍ１を用いずに、評価値と基準値との比較によりバリの状態を予測する点で、第１実施形態に係る成形システム１０と相違する。 FIG. 12 is a block diagram schematically showing the molding system 11 according to the second embodiment. The molding system 11 includes a plurality of molding devices 20, an abnormality prediction device 40a, an input unit 50, and a display unit 60. In the present embodiment, the abnormality prediction device 40a of the molding system 11 acquires prediction information for predicting the burr state of the molded product based on the evaluation value and a predetermined reference value. That is, the molding system 11 is different from the molding system 10 according to the first embodiment in that the state of burrs is predicted by comparing the evaluation value and the reference value without using the trained model Tm1.
　図１３は、本実施形態に係る異常予測装置４０ａの機能構成を示すブロック図である。異常予測装置４０ａは、データ取得部４１と、異常予測部４２ａと、出力部４３ａと、成形情報記憶部４４と、基準値記憶部４６とを有する。これらの各部は、ＣＰＵ等の演算部とＨＤＤ等の記憶部とを有するコンピュータ装置により実現される。データ取得部４１は、第１実施形態に係るデータ取得部４１と同様に、評価情報を取得する。 FIG. 13 is a block diagram showing a functional configuration of the abnormality prediction device 40a according to the present embodiment. The abnormality prediction device 40a includes a data acquisition unit 41, an abnormality prediction 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 an HDD. The data acquisition unit 41 acquires evaluation information in the same manner as the data acquisition unit 41 according to the first embodiment.
　基準値記憶部４６には、複数の基準値が記憶されている。基準値は、成型品が正常に成形された際に取得される評価値に基づいて生成される値である。基準値は、例えば、複数の成型品の正常成形時にそれぞれ取得される複数の評価値の平均値又は中央値に、所定のマージンを加えた値である。所定のマージンは、成型品において許容されるバリの最大長さ等により決定される。 A plurality of reference values are stored in the reference value storage unit 46. The reference value is a value generated based on the evaluation value obtained when the molded product is normally molded. The reference value is, for example, a value obtained by adding a predetermined margin to the average value or the median value of a plurality of evaluation values obtained during normal molding of a plurality of molded products. The predetermined margin is determined by the maximum length of burrs allowed in the molded product and the like.
　基準値は、第２評価値ごとに、第３評価値ごとに、又は成形情報ごとに複数の値が用意されている。例えば、第２評価値（温度に関する値）が第１値未満のときには第１基準値、第２評価値が第１値以上第２値未満のときには第２基準値、第２評価値が第２値以上のときには第３基準値が用いられるようにしてもよい。基準値記憶部４６には、例えば、第２評価値、第３評価値及び成形情報と、複数の基準値とを対応付けしたテーブル形式の情報が記憶されている。 As the reference value, a plurality of values are prepared for each of the second evaluation values, for each of the third evaluation values, or for each molding information. For example, when the second evaluation value (value related to temperature) is less than the first value, the first reference value is used, when the second evaluation value is equal to or more than the first value and less than the second value, the second reference value and the second evaluation value are the second. When it is equal to or more than the value, the third reference value may be used. The reference value storage unit 46 stores, for example, table-type information in which the second evaluation value, the third evaluation value, and the molding information are associated with a plurality of reference values.
　異常予測部４２ａは、データ取得部４１により取得された第２評価値、第３評価値及び成形情報に基づいて、基準値記憶部４６に記憶されている複数の基準値から対応する基準値（以下、「注目基準値」と称する。）を取得する。そして、異常予測部４２ａは、データ取得部４１により取得された評価値（すなわち、予測対象となる成型品の成形時に取得される評価値）と、注目基準値とを比較することで、予測情報を取得する。予測情報は、例えば、評価値と注目基準値との差、又は比である。 The abnormality prediction unit 42a has a reference value (corresponding from a plurality of reference values stored in the reference value storage unit 46, based on the second evaluation value, the third evaluation value, and the molding information acquired by the data acquisition unit 41. Hereinafter, it is referred to as “attention reference value”). Then, the abnormality prediction unit 42a compares the evaluation value acquired by the data acquisition unit 41 (that is, the evaluation value acquired at the time of molding the molded product to be predicted) with the attention reference value to predict the prediction information. To get. The prediction information is, for example, the difference or ratio between the evaluation value and the attention reference value.
　図６から図８（ｂ）において説明したように、バリが発生する場合、評価値（例えば、圧力の傾き）は、注目基準値よりも大きくなる傾向がある。このため、注目基準値よりも評価値が大きくなる場合、予測対象となる成型品にバリが発生していることが予測される。したがって、予測情報には、予測対象となる成型品にバリが発生しているか否かが示されている。 As described in FIGS. 6 to 8 (b), when burrs occur, the evaluation value (for example, the slope of pressure) tends to be larger than the reference value of interest. Therefore, when the evaluation value is larger than the attention reference value, it is predicted that burrs are generated in the molded product to be predicted. Therefore, the prediction information indicates whether or not burrs are generated in the molded product to be predicted.
　図１４は、異常予測部４２ａにおける評価値と注目基準値との比較を模式的に説明するグラフである。図１４において、縦軸は圧力の傾き（ｄＰ／ｄｔ）であり、横軸は時間ｔである。また、図１４には注目基準値Ｔｈ１を併せて示している。グラフ線Ｆ３４、Ｆ３５は、それぞれ予測対象となる成型品を成形した際に圧力センサ２５により得られる圧力の時系列データに基づいて、データ取得部４１により取得された圧力の傾きの時系列データである。 FIG. 14 is a graph schematically explaining the comparison between the evaluation value and the attention reference value in the abnormality prediction unit 42a. In FIG. 14, the vertical axis is the slope of pressure (dP / dt), and the horizontal axis is time t. In addition, FIG. 14 also shows the attention reference value Th1. The graph lines F34 and F35 are time-series data of the pressure gradient acquired by the data acquisition unit 41 based on the time-series data of the pressure obtained by the pressure sensor 25 when the molded product to be predicted is molded, respectively. be.
　評価値が圧力の傾きの最大値である場合、図１４の例では評価値はｄＰ４、ｄＰ５となる。図１４の例では、注目基準値Ｔｈ１よりも評価値ｄＰ４の方が大きいため、グラフ線Ｆ３４が取得された成型品には、バリが発生していることが予測される。一方、注目基準値Ｔｈ１よりも評価値ｄＰ５の方が小さいため、グラフ線Ｆ３５が取得された成型品には、バリが発生していない（又は、バリが発生しているとしても許容範囲内である）ことが予測される。 When the evaluation value is the maximum value of the pressure gradient, the evaluation values are dP4 and dP5 in the example of FIG. In the example of FIG. 14, since the evaluation value dP4 is larger than the attention reference value Th1, it is predicted that burrs are generated in the molded product from which the graph line F34 is acquired. On the other hand, since the evaluation value dP5 is smaller than the attention reference value Th1, no burrs are generated in the molded product from which the graph line F35 is acquired (or even if burrs are generated, they are within the permissible range. Yes) is expected.
　図１３を参照する。出力部４３ａは、異常予測部４２ａにおいて取得された予測情報に基づいて、成型品のバリの状態の良否を判定する。例えば、予測情報が評価値と注目基準値との差（評価値－注目基準値）である場合、出力部４３ａは、予測情報が正の値である場合に、成型品のバリの状態が不良（異常）であると判定する。 Refer to FIG. The output unit 43a determines whether or not the burr state of the molded product is good or bad based on the prediction information acquired by the abnormality prediction unit 42a. For example, when the prediction information is the difference between the evaluation value and the attention reference value (evaluation value-attention reference value), the output unit 43a has a poor burr state of the molded product when the prediction information is a positive value. Judged as (abnormal).
　出力部４３ａは、成型品のバリの良否に関する判定情報と、予測情報とを表示部６０及び制御部２７１に出力する。表示部６０及び制御部２７１は、判定情報及び予測情報に基づいて、第１実施形態と同様の動作を行う。 The output unit 43a outputs the determination information regarding the quality of the burr of the molded product and the prediction information to the display unit 60 and the control unit 271. The display unit 60 and the control unit 271 perform the same operations as in the first embodiment based on the determination information and the prediction information.
　なお、本実施形態において、出力部４３ａを設けずに、異常予測部４２ａにおいて得られた予測情報をそのまま表示部６０に表示するように構成されてもよい。この場合、表示部６０に表示された予測情報に基づいて、オペレータが成型品の良否や、成形装置２０の状態を判断するようにしてもよい。 In the present embodiment, the prediction information obtained by the abnormality prediction unit 42a may be displayed as it is on the display unit 60 without providing the output unit 43a. In this case, the operator may determine the quality of the molded product and the state of the molding apparatus 20 based on the prediction information displayed on the display unit 60.
　本実施形態に係る成形システム１１によれば、基準値と評価値との比較により、バリの状態を容易に予測することができる。 According to the molding system 11 according to the present embodiment, the state of burrs can be easily predicted by comparing the reference value and the evaluation value.
　以上のとおり開示した実施形態はすべての点で例示であって制限的なものではない。つまり、本開示の成形システムは、図示する形態に限られず、本開示の範囲内において他の形態であってもよい。 The embodiments disclosed as described above are examples in all respects and are not restrictive. That is, the molding system of the present disclosure is not limited to the illustrated form, and may be another form within the scope of the present disclosure.
　本出願は、２０２０年４月７日出願の日本特許出願（特願２０２０－０６９２９０）に基づくものであり、その内容はここに参照として取り込まれる。 This application is based on a Japanese patent application filed on April 7, 2020 (Japanese Patent Application No. 2020-069290), the contents of which are incorporated herein by reference.
　１０、１１　成形システム
　２０　成形装置　　　　　　　　２１　ベッド
　２２　射出部　　　　　　　　２２１　ホッパ　　　　　２２２　シリンダ
　２２３　スクリュ　　　　　　２２４　ノズル　　　　　２２５　ボールねじ
　２２６　モータ　　　　　　　２２７　与圧センサ　　　２２８　ヒータ
　２３　型締め部　　　　　　　２３１　固定盤　　　　　２３２　可動盤
　２３３　タイバー　　　　　　２３４　ボールねじ　　　２３５　支持盤
　２３６　型締め力センサ　　　２３７　モータ　　　　　　２４　金型部
　２４１　金型　　　　　　　２４１ａ　露出面　　　　　２４２　金型
　２４３　流路　　　　　　　　　２５　圧力センサ　　　　２６　温度センサ
　２７　制御盤　　　　　　　　２７１　制御部　　　　　２７２　通信部
　３０　学習装置　　　　　　　　３１　訓練データ取得部　３２　学習演算部
　３３　成形情報記憶部　　　　　３４　学習済みモデル記憶部
　４０、４０ａ　異常予測装置　　４１　データ取得部
　４２、４２ａ　異常予測部　　　　　　　　　　　４３、４３ａ　出力部
　４４　成形情報記憶部　　　　　４５　学習済みモデル記憶部
　４６　基準値記憶部　　　　　　５０　入力部　　　　　　６０　表示部
　Ｃ１　キャビティ　　　　　　Ｃ１ａ　領域　　　　　　　Ｃ２　隙間
　Ｌ１　成形材料　　　　　　　Ｆｍ１　異物　　　　　　 10, 11 Molding system 20 Molding equipment 21 Bed 22 Injection part 221 Hopper 222 Cylinder 223 Screw 224 Nozzle 225 Ball screw 226 Motor 227 Pressure sensor 228 Heater 23 Type tightening part 231 Fixed plate 232 Movable plate 233 Tie bar 234 Ball screw 235 Support plate 236 Mold tightening force sensor 237 Motor 24 Mold part 241 Mold 241a Exposed surface 242 Mold 243 Flow path 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 Learned model storage unit 40, 40a Abnormality prediction device 41 Data acquisition unit 42, 42a Abnormality prediction unit 43, 43a Output unit 44 Molding information storage unit 45 Learned model storage unit 46 Reference value storage unit 50 Input Part 60 Display part C1 Cavity C1a Area C2 Gap L1 Molding material Fm1 Foreign matter

Claims (10)

  1. 　成形装置と、
    　前記成形装置により成形される成型品の異常を予測する異常予測装置と、
    を備える成形システムであって、
    　前記成形装置は、
    　　複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め部と、
    　　前記複数の金型を締め付けることで前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填動作と、前記キャビティに充填された前記成形材料の圧力を保持する保圧動作と、前記成形材料の圧力の保持を解除する保圧解除動作と、を行う射出部と、
    　　前記キャビティ中の前記成形材料の圧力を検出する圧力センサと、
    を有し、
    　前記異常予測装置は、
    　　前記圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除動作後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得部と、
    　　前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測部と、
    を有する、成形システム。     Molding equipment and
    An abnormality prediction device that predicts an abnormality in a molded product molded by the molding device,
    It is a molding system equipped with
    The molding apparatus is
    A mold clamping part that tightens the plurality of molds in a state where a plurality of molds are combined,
    A filling operation of filling a cavity formed between the plurality of molds with a molten molding material by tightening the plurality of molds, and a holding pressure for holding the pressure of the molding material filled in the cavity. An injection unit that performs an operation and a pressure holding release operation for releasing the pressure holding of the molding material.
    A pressure sensor that detects the pressure of the molding material in the cavity, and
    Have,
    The abnormality prediction device is
    Based on the time-series data of the pressure detected by the pressure sensor, a data acquisition unit that acquires an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation, and a data acquisition unit.
    An abnormality prediction unit that acquires prediction information for predicting the burr state of the molded product based on the evaluation value, and
    Has a molding system.
  2. 　前記異常予測部は、前記評価値と前記成型品のバリの状態との相関関係を機械学習させた学習済みモデルへ前記評価値を入力することで、前記予測情報を取得し、
    　前記学習済みモデルの説明変数は、前記評価値を含み、
    　前記学習済みモデルの目的変数は、前記成型品のバリの状態であり、
    　前記バリの状態は、前記バリの有無、又は前記バリの大きさを含む、
    請求項１に記載の成形システム。     The abnormality prediction unit acquires the prediction information by inputting the evaluation value into a trained model in which the correlation between the evaluation value and the burr state of the molded product is machine-learned.
    The explanatory variables of the trained model include the evaluation values.
    The objective variable of the trained model is the burr state of the molded product.
    The state of the burr includes the presence or absence of the burr or the size of the burr.
    The molding system according to claim 1.
  3. 　前記成形装置は、前記成形材料の温度を検出する温度センサをさらに有し、
    　前記学習済みモデルの前記説明変数は、前記温度センサにより検出された温度に関する第２評価値を含み、
    　前記異常予測部は、前記学習済みモデルへ前記評価値及び前記第２評価値を入力することで、前記予測情報を取得する、
    請求項２に記載の成形システム。     The molding apparatus further includes a temperature sensor that detects the temperature of the molding material.
    The explanatory variables of the trained model include a second evaluation value for the temperature detected by the temperature sensor.
    The abnormality prediction unit acquires the prediction information by inputting the evaluation value and the second evaluation value into the trained model.
    The molding system according to claim 2.
  4. 　前記異常予測部は、前記成型品の正常成形時に取得される前記評価値に基づいて生成される基準値よりも、予測対象となる前記成型品の成形時に取得される前記評価値が大きくなる場合に、前記予測対象となる前記成型品にバリが発生していることを示す前記予測情報を取得する、
    請求項１に記載の成形システム。     When the abnormality prediction unit obtains a larger evaluation value during molding of the molded product to be predicted than a reference value generated based on the evaluation value acquired during normal molding of the molded product. In addition, the prediction information indicating that burrs are generated in the molded product to be predicted is acquired.
    The molding system according to claim 1.
  5. 　前記データ取得部は、前記保圧解除動作後の前記時系列データに基づいて得られる圧力の傾きの第２時系列データに基づいて、前記評価値を取得し、
    　前記評価値は、前記第２時系列データにおける前記傾きの最大値、又は前記第２時系列データのうち前記傾きが０以上となる領域の幅である、
    請求項１から請求項４のいずれか１項に記載の成形システム。     The data acquisition unit acquires the evaluation value based on the second time-series data of the pressure gradient obtained based on the time-series data after the pressure holding release operation.
    The evaluation value is the maximum value of the slope in the second time series data, or the width of the region of the second time series data in which the slope is 0 or more.
    The molding system according to any one of claims 1 to 4.
  6. 　前記データ取得部は、前記保圧解除動作後の前記時系列データの極小における圧力と極大における圧力との差、又は前記極小における圧力から前記極大における圧力への変化率を、前記評価値として取得する、
    請求項１から請求項４のいずれか１項に記載の成形システム。     The data acquisition unit acquires, as the evaluation value, the difference between the pressure at the minimum and the pressure at the maximum, or the rate of change from the pressure at the minimum to the pressure at the maximum after the pressure holding release operation. do,
    The molding system according to any one of claims 1 to 4.
  7. 　成形装置により成形される成型品の異常を予測する異常予測装置であって、
    　前記成形装置は、
    　　複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め部と、
    　　前記複数の金型を締め付けることで前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填動作と、前記キャビティに充填された前記成形材料の圧力を保持する保圧動作と、前記成形材料の圧力の保持を解除する保圧解除動作と、を行う射出部と、
    　　前記キャビティ中の前記成形材料の圧力を検出する圧力センサと、
    を有し、
    　前記異常予測装置は、
    　　前記圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除動作後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得部と、
    　　前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測部と、
    を備える、異常予測装置。     An abnormality prediction device that predicts abnormalities in a molded product molded by a molding device.
    The molding apparatus is
    A mold clamping part that tightens the plurality of molds in a state where a plurality of molds are combined,
    A filling operation of filling a cavity formed between the plurality of molds with a molten molding material by tightening the plurality of molds, and a holding pressure for holding the pressure of the molding material filled in the cavity. An injection unit that performs an operation and a pressure holding release operation for releasing the pressure holding of the molding material.
    A pressure sensor that detects the pressure of the molding material in the cavity, and
    Have,
    The abnormality prediction device is
    Based on the time-series data of the pressure detected by the pressure sensor, a data acquisition unit that acquires an evaluation value regarding a change in pressure of the molding material after the pressure holding release operation, and a data acquisition unit.
    An abnormality prediction unit that acquires prediction information for predicting the burr state of the molded product based on the evaluation value, and
    An abnormality predictor.
  8. 　複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測する異常予測方法であって、
    　前記キャビティ中の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除工程後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得工程と、
    　前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測工程と、
    を備える、異常予測方法。     A mold clamping step of tightening the plurality of molds in a state where a plurality of molds are combined, and a filling step of filling a cavity formed between the plurality of molds by the mold clamping step with a molten molding material. An abnormality of a molded product molded by a molding method including a pressure holding step of holding the pressure of the molding material filled in the cavity and a pressure holding release step of releasing the pressure holding of the molding material. It is an abnormality prediction method to predict
    A data acquisition step of acquiring an evaluation value regarding a change in the pressure of the molding material after the pressure holding release step based on the time-series data of the pressure detected by the pressure sensor that detects the pressure of the molding material in the cavity. When,
    An abnormality prediction step for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value, and
    An abnormality prediction method.
  9. 　複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測するためのプログラムであって、
    　前記キャビティ中の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて、前記保圧解除工程後の前記成形材料の圧力の変化に関する評価値を取得するデータ取得処理と、
    　前記評価値に基づいて、前記成型品のバリの状態を予測するための予測情報を取得する異常予測処理と、
    をコンピュータ装置に実行させる、プログラム。     A mold clamping step of tightening the plurality of molds in a state where a plurality of molds are combined, and a filling step of filling a cavity formed between the plurality of molds by the mold clamping step with a molten molding material. An abnormality of a molded product molded by a molding method including a pressure holding step of holding the pressure of the molding material filled in the cavity and a pressure holding release step of releasing the pressure holding of the molding material. It ’s a prediction program.
    A data acquisition process for acquiring an evaluation value regarding a change in the pressure of the molding material after the pressure holding release step based on the time-series data of the pressure detected by the pressure sensor that detects the pressure of the molding material in the cavity. When,
    Anomaly prediction processing for acquiring prediction information for predicting the burr state of the molded product based on the evaluation value, and
    A program that causes a computer device to execute.
  10. 　複数の金型を組み合わせた状態で、前記複数の金型を締め付ける型締め工程と、前記型締め工程により前記複数の金型の間に形成されるキャビティへ溶融状態の成形材料を充填する充填工程と、前記キャビティに充填された前記成形材料の圧力を保持する保圧工程と、前記成形材料の圧力の保持を解除する保圧解除工程と、を備える成形方法により成形される成型品の異常を予測するための学習済みモデルであって、
    　説明変数は、前記キャビティ中の前記成形材料の圧力を検出する圧力センサにより検出された圧力の時系列データに基づいて取得される、前記保圧解除動作後の前記成形材料からの圧力の変化に関する評価値を含み、
    　目的変数は、前記成型品のバリの状態であり、
    　前記バリの状態は、前記バリの有無、又は前記バリの大きさを含む、
    学習済みモデル。
          A mold clamping step of tightening the plurality of molds in a state where a plurality of molds are combined, and a filling step of filling a cavity formed between the plurality of molds by the mold clamping step with a molten molding material. An abnormality of a molded product molded by a molding method including a pressure holding step of holding the pressure of the molding material filled in the cavity and a pressure holding release step of releasing the pressure holding of the molding material. A trained model for predicting
    The explanatory variables relate to the change in pressure from the molding material after the pressure holding release operation, which is acquired based on the time series data of the pressure detected by the pressure sensor that detects the pressure of the molding material in the cavity. Including evaluation value
    The objective variable is the burr state of the molded product.
    The state of the burr includes the presence or absence of the burr or the size of the burr.
    Trained model.
PCT/JP2021/014318 2020-04-07 2021-04-02 Molding system, irregularity prediction device, irregularity prediction method, program, and trained model WO2021206017A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2020069290A JP2021165014A (en) 2020-04-07 2020-04-07 Molding system, abnormality prediction device, abnormality prediction method, program and learned model
JP2020-069290 2020-04-07

Publications (1)

Publication Number Publication Date
WO2021206017A1 true WO2021206017A1 (en) 2021-10-14

Family

ID=78021575

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/014318 WO2021206017A1 (en) 2020-04-07 2021-04-02 Molding system, irregularity prediction device, irregularity prediction method, program, and trained model

Country Status (2)

Country Link
JP (2) JP2021165014A (en)
WO (1) WO2021206017A1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07290548A (en) * 1994-04-22 1995-11-07 Matsushita Electric Works Ltd Method for judging quality of molded product of injection molding machine
JP2000052396A (en) * 1998-08-12 2000-02-22 Rika Kogyo Kk Device and method for controlling injection molding
JP2009137076A (en) * 2007-12-04 2009-06-25 Autonetworks Technologies Ltd Injection molding mold, method for detecting defective plasticization in injection molding, and injection molding method
WO2015186417A1 (en) * 2014-06-04 2015-12-10 オリンパス株式会社 Injection molding method for resin molding and method for specifying mold-clamping force
JP2015231676A (en) * 2014-06-09 2015-12-24 株式会社日本製鋼所 Molding quality determination method for hollow molding
JP2020049929A (en) * 2018-09-20 2020-04-02 株式会社ジェイテクト Molding conditions decision support device and injection molding machine

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07290548A (en) * 1994-04-22 1995-11-07 Matsushita Electric Works Ltd Method for judging quality of molded product of injection molding machine
JP2000052396A (en) * 1998-08-12 2000-02-22 Rika Kogyo Kk Device and method for controlling injection molding
JP2009137076A (en) * 2007-12-04 2009-06-25 Autonetworks Technologies Ltd Injection molding mold, method for detecting defective plasticization in injection molding, and injection molding method
WO2015186417A1 (en) * 2014-06-04 2015-12-10 オリンパス株式会社 Injection molding method for resin molding and method for specifying mold-clamping force
JP2015231676A (en) * 2014-06-09 2015-12-24 株式会社日本製鋼所 Molding quality determination method for hollow molding
JP2020049929A (en) * 2018-09-20 2020-04-02 株式会社ジェイテクト Molding conditions decision support device and injection molding machine

Also Published As

Publication number Publication date
JP2021165014A (en) 2021-10-14
JP2024051055A (en) 2024-04-10

Similar Documents

Publication Publication Date Title
US20200103862A1 (en) Deterioration determining apparatus and deterioration determining system
US20170001356A1 (en) Method of injection molding with constant-velocity flow front control
US20080099943A1 (en) Injection molding control method
US11358317B2 (en) Systems and methods for controlling injection molding using predicted cavity pressure
JP7159758B2 (en) Molding condition determination support device and injection molding machine
US11537945B2 (en) Machine learning device and machine learning method
WO2021206017A1 (en) Molding system, irregularity prediction device, irregularity prediction method, program, and trained model
WO2021215222A1 (en) Molding system, anomaly prediction device, anomaly prediction method, program, and trained model
Huang et al. Quality monitoring of micro-shrinkage defects in thick-walled injection molded components
WO2021220719A1 (en) Molding system, abnormality prediction device, abnormality prediction method, program, and learned model
JP2006289773A (en) Device, method and program for monitoring molding machine
WO2019194974A1 (en) Method of determining melt flow front travel using injection melt pressure and screw position to generate an output and continuously monitor check ring repeatability
JP2022077628A (en) Abnormality prediction system, abnormality prediction device, abnormality prediction method, program, and learned model
JP7293962B2 (en) Molded product quality anomaly prediction system
US20200206998A1 (en) Quality prediction system and molding machine
JP5419364B2 (en) Injection molding system and injection molding method
JP6660443B2 (en) Injection molding apparatus and injection molding method
Minh et al. A study on the welding line strength of composite parts with various venting systems in injection molding process
JP2009137076A (en) Injection molding mold, method for detecting defective plasticization in injection molding, and injection molding method
JP7424192B2 (en) Molding system, abnormality prediction device, abnormality prediction method, program and learned model
JP2022088931A (en) Anomaly detection system, anomaly detection device, anomaly detection method, program and learned model
JP7481048B1 (en) Information processing system, information processing method, and program
US20200094461A1 (en) Device for assisting molding condition determination and injection molding apparatus
JP2022098299A (en) Detection system, detection device, detection method, program, and learned model
JP2022092311A (en) Abnormality detection system, abnormality detection device, abnormality detection method, program and learnt model

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21784098

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21784098

Country of ref document: EP

Kind code of ref document: A1