WO2022210425A1 - 射出成形条件の探索支援装置及びその方法、プログラム及び記録媒体 - Google Patents
射出成形条件の探索支援装置及びその方法、プログラム及び記録媒体 Download PDFInfo
- Publication number
- WO2022210425A1 WO2022210425A1 PCT/JP2022/014669 JP2022014669W WO2022210425A1 WO 2022210425 A1 WO2022210425 A1 WO 2022210425A1 JP 2022014669 W JP2022014669 W JP 2022014669W WO 2022210425 A1 WO2022210425 A1 WO 2022210425A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- injection molding
- quality
- probability
- molding condition
- molding conditions
- Prior art date
Links
- 238000001746 injection moulding Methods 0.000 title claims abstract description 226
- 238000000034 method Methods 0.000 title claims description 57
- 238000003860 storage Methods 0.000 title description 12
- 238000009826 distribution Methods 0.000 claims abstract description 88
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 230000008859 change Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 74
- 238000000465 moulding Methods 0.000 claims description 48
- 230000008569 process Effects 0.000 claims description 41
- 238000002347 injection Methods 0.000 claims description 28
- 239000007924 injection Substances 0.000 claims description 28
- 238000013145 classification model Methods 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 4
- 238000007477 logistic regression Methods 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 2
- 230000006872 improvement Effects 0.000 claims description 2
- 239000000463 material Substances 0.000 description 21
- 230000007246 mechanism Effects 0.000 description 9
- 238000009795 derivation Methods 0.000 description 8
- 238000001816 cooling Methods 0.000 description 7
- 238000005429 filling process Methods 0.000 description 7
- 238000005303 weighing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 230000000875 corresponding effect Effects 0.000 description 4
- 238000013500 data storage Methods 0.000 description 4
- 230000002950 deficient Effects 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- 239000008188 pellet Substances 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000011347 resin Substances 0.000 description 2
- 229920005989 resin Polymers 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004014 plasticizer Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000007711 solidification Methods 0.000 description 1
- 230000008023 solidification Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
- B29C45/76—Measuring, controlling or regulating
- B29C45/766—Measuring, controlling or regulating the setting or resetting of moulding conditions, e.g. before starting a cycle
-
- 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
- B29C2945/00—Indexing scheme relating to injection moulding, i.e. forcing the required volume of moulding material through a nozzle into a closed mould
- B29C2945/76—Measuring, controlling or regulating
- B29C2945/76929—Controlling method
- B29C2945/76939—Using stored or historical data sets
- B29C2945/76949—Using stored or historical data sets using a learning system, i.e. the system accumulates experience from previous occurrences, e.g. adaptive control
Definitions
- the present disclosure relates to an injection molding condition search support device, its method, program, and recording medium.
- Patent Document 1 relates to an injection molding machine with a machine learning function, and in particular discloses a device and method for supporting determination of molding conditions. Specifically, it learns the correlation between the molding conditions and the quality of the molded product, and when a quality-related defect occurs, the user is presented with which molding conditions should be adjusted in order to improve the quality defect. See paragraphs 0009 to 0010 of the same document).
- the technology for obtaining good product conditions by modeling the relationship between injection molding conditions and molding quality using machine learning assumes a well-learned model.
- the accuracy may decrease because the influence of noise on molding quality cannot be considered.
- a situation where there is not enough learning data is, for example, the launch of a new mold.
- Different molds have different relationships between injection molding conditions and molding quality, so even if there is existing learning data, it cannot be used as it is.
- Noise related to molding quality is considered to be small for molding quality measured by various sensors, but tends to be large for quality evaluated by humans.
- the inventors of the present application have developed an apparatus and method for assisting the search for injection molding conditions that can facilitate reducing or avoiding the effects of noise on molding quality and/or restrictions on the number of learning data. I found a new meaning in providing.
- An injection molding condition search support device is an injection molding condition search support device for an injection molding machine that manufactures a molded product by executing injection molding based on input of injection molding conditions, A prediction model generation unit that generates a prediction model for predicting quality regarding unknown injection molding conditions based on prior data in which injection molding condition data and molded product quality data are associated, wherein the prediction model includes (i) injection molding A prediction model for calculating a first probability distribution in which the predicted value and variance of quality continuously change according to changes in conditions, and/or (ii) the occurrence of specific phenomena related to quality according to changes in injection molding conditions.
- a predictive model generation unit including a predictive model for calculating a second probability distribution whose probability changes continuously; Evaluation of one or more first probability distributions on quality and/or evaluation of two or more second probability distributions on two or more specific phenomena and/or at least one first probability distribution and at least one second probability
- An injection molding condition determination unit is included that determines the next injection molding condition to be set in the injection molding machine based on the evaluation of the distribution.
- An injection molding condition search support method is an injection molding condition search support method for an injection molding machine that manufactures a molded product by executing injection molding based on input of injection molding conditions, Generating a predictive model that predicts quality for unknown injection molding conditions based on prior data associated with injection molding condition data and molded product quality data, the predictive model: (i) in response to changes in the injection molding conditions; A prediction model for calculating a first probability distribution in which the predicted value and variance of quality change continuously, and/or (ii) the probability of occurrence of a specific phenomenon related to quality changes continuously according to changes in injection molding conditions.
- a prediction model for calculating a second probability distribution to Determining the next injection molding condition to be set in the injection molding machine based on the evaluation of one or more first probability distributions regarding quality and/or the evaluation of two or more second probability distributions regarding two or more specific phenomena. do.
- a program for implementing this method is similarly comprehensible, and the concept is substantially disclosed in this specification.
- the program can be distributed by being recorded on a non-temporary recording medium (for example, an optical disk, a magnetic disk, a hard disk, a semiconductor memory, etc.) as well as being downloaded from a server.
- a non-transitory storage medium is a tangible object that does not contain a communication line on which such programs are temporarily propagated as data.
- an apparatus and method for assisting in the search for injection molding conditions that can facilitate reducing or avoiding the effects of constraints on the number of training data and/or noise on molding quality.
- FIG. 1 is a schematic diagram showing a schematic configuration of an injection molding machine according to one aspect of the present disclosure
- FIG. 2 is a schematic block diagram mainly showing a control section of the injection molding machine
- FIG. 2 is a schematic block diagram mainly showing a predictive model generator and an injection molding condition determiner
- It is a schematic diagram which shows a 1st probability distribution.
- FIG. 10 is a schematic diagram showing two different second probability distributions derived by the classification model derivation unit for each of two different qualities
- 4 is a schematic flow chart showing the operation of the control section of the injection molding machine;
- Non-limiting embodiments and features of the present invention will now be described with reference to FIGS.
- a person skilled in the art can combine each embodiment and/or each feature without undue explanation, and can also understand the synergistic effect of this combination. Redundant explanations among the embodiments will be omitted in principle.
- the referenced drawings are primarily for the purpose of describing the invention and are simplified for drawing convenience. Each feature is not valid only for the search support device disclosed in this specification, but is understood as a universal feature that applies to various other search support devices not disclosed in this specification. .
- an injection molding machine 1 has a mold clamping device 2 and an injection device 3 mounted on a common or different base 4 .
- the injection molding machine 1 continuously manufactures molded products based on cooperative operations of the mold clamping device 2 and the injection device 3 .
- the mold clamping device 2 is configured to repeat a mold closing, mold clamping, and mold opening loop.
- the injection device 3 is configured to repeat a loop of a metering process, a filling process, and a holding pressure process.
- a mold device 5 is attached to the mold clamping device 2 .
- a specific configuration of the mold device 5 is determined by the shape, size, and number of injection-molded articles.
- the mold device 5 can be of the two-plate or three-plate type. In some forms, mold assembly 5 has one or more stationary molds 51 and one or more movable molds 52 .
- the mold clamping device 2 has a stationary platen 21 , a movable platen 22 , a toggle mechanism 23 , a toggle support 24 , a plurality of tie bars 25 , a mold clamping motor 26 and a mold thickness adjusting mechanism 27 .
- the toggle support 24 and the movable platen 22 are connected via the toggle mechanism 23 , and the movable platen 22 can move forward and backward with respect to the stationary platen 21 based on the operation of the toggle mechanism 23 .
- the operation of the mold clamping motor 26 changes the state of the toggle mechanism 23 and changes the position of the movable platen 22 .
- the mold device 5 can be introduced into the space between the stationary platen 21 and the movable platen 22 .
- fixed and movable molds 51, 52 are attached to fixed and movable platens 21, 22, respectively.
- the movable platen 22 is moved toward the stationary platen 21 to close the mold device 5, then clamp the mold, and finally open the mold.
- the mold is closed in a state in which the opposing surface of the fixed mold 51 and the opposing surface of the movable mold 52 are in contact with each other and the half cavity of the fixed mold 51 and the half cavity of the movable mold 52 are in spatial communication. be.
- Mold clamping is a state in which the movable mold 52 is strongly pressed by the fixed mold 51 in order to withstand the injection pressure of the material from the injection device 3 .
- Mold opening is a state in which the facing surface of the fixed mold 51 and the facing surface of the movable mold 52 are not in contact with each other and a space is provided between them.
- the toggle mechanism 23 includes a crosshead 23a that receives a driving force from a mold clamping motor 26, first and second links 23b and 23c that are pivotably connected between the toggle support 24 and the movable platen 22, the crosshead 23a and the first link 23c. It has a third link 23d connecting between the links 23b.
- the rotational force generated by the mold clamping motor 26 is converted into linear thrust by a force transducer such as a ball screw 262 via a belt 261 and applied to the crosshead 23a. For example, when the output shaft of the mold clamping motor 26 rotates forward, the crosshead 23a moves straight toward the fixed platen 21, the angle formed by the first link 23b and the second link 23c increases, and the movable platen 22 moves toward the fixed platen.
- the crosshead 23a In response to the reverse rotation of the output shaft of the mold clamping motor 26, the crosshead 23a is moved away from the fixed platen 21, the angle formed by the first link 23b and the second link 23c becomes smaller, and the movable platen 22 moves toward the fixed platen. Go straight away from 21.
- the direction in which the movable platen 22 and the movable mold 52 attached thereto move toward the stationary platen 21 and the stationary mold 51 attached thereto is defined as the front side or the injection device side, This opposite direction can also be defined as the rear side or the reflection molding device side.
- the toggle mechanism 23 operates to double the thrust applied to the crosshead 23 a and transmit it to the movable platen 22 .
- the magnification is also called toggle magnification.
- the toggle magnification changes according to the angle formed by the first link 23b and the second link 23c. As the angle formed by the first link 23b and the second link 23c approaches 180°, the toggle magnification also increases.
- the mold thickness adjustment mechanism 27 is configured to adjust the position of the toggle support 24 with respect to the stationary platen 21 (the front-to-rear distance between the two, so to speak, the mold thickness).
- the mold thickness adjusting mechanism 27 includes a mold thickness adjusting motor 27a.
- the rotational force generated by the mold thickness adjusting motor 27a is transmitted to the nut screwed on the screw shaft at the rear end of the tie bar 25 via the belt 271, changing the position of the toggle support 24 along the tie bar 25 and fixing it.
- the position of the toggle support 24 with respect to the platen 21 (that is, the spacing therebetween) is changed.
- the rotational force of the mold thickness adjusting motor 27a is transmitted to the nut via transmission elements such as belts and gears (or directly).
- the mold clamping device 2 includes an ejector device 28 for ejecting the molded product from the mold device 5.
- the ejector device 28 is attached behind the movable platen 22, for example.
- the ejector device 28 includes an ejector rod and an ejector motor that powers the ejector rod.
- the rotational force generated by the ejector motor is converted into linear force by the ball screw and transmitted to the ejector rod.
- the ejector rod When the ejector rod is advanced, it pushes the ejector plate of the mold device 5 .
- the molded product of the movable mold 52 is pushed by the ejector pin and ejected from the mold device 5 .
- the injection molding machine 1 operates an ejector device in synchronization with mold opening.
- the injection device 3 supplies molten resin material to the mold device 5 attached to the mold clamping device 2 .
- the injection device can be of the in-line screw type or pre-plasticizer type. In this specification, the injection device is described as an in-line screw type, but it should not be limited to this.
- the injection device 3 has a cylinder 31 , a screw 32 , a heater 33 , a metering motor 34 , an injection motor 35 , a moving motor 36 , a guide rail 37 , a first movable support 38 and a second movable support 39 .
- the cylinder 31 is a metal cylindrical member that accommodates the screw 32, and has a cylinder body portion 31a and a nozzle portion 31b.
- the cylinder body 31a accommodates the screw 32 .
- the nozzle portion 31b has a straight channel with a channel diameter smaller than that of the cylinder body 31a, and has a discharge port for discharging the molten plastic material supplied from the cylinder body 31a.
- the cylinder body 31a has a material supply port 31c for receiving plastic material, such as pellets, supplied from a hopper 31f or an automated plastic material supply device. The pellets are melted by the heat transmitted from the heater 33 through the cylinder body 31a, and conveyed toward the front side, that is, toward the nozzle section 31b as the screw 32 rotates.
- the moving direction of the screw 32 during filling is the front side
- the moving direction of the screw 32 during metering is the rear side.
- the screw 32 has a shaft portion and a spiral flight provided on the outer circumference of the shaft portion, and conveys solid and molten resin materials to the front side of the cylinder 31 according to its rotation.
- the screw 32 can be rotated by receiving rotational force from the metering motor 34 .
- the output shaft of the weighing motor 34 and the screw 32 are mechanically connected via the belt 341 .
- the screw 32 can receive a driving force from the injection motor 35 and move forward (to approach the nozzle portion 31b) and rearward (to move away from the nozzle portion 31b) within the stationary cylinder 31 .
- the output shaft of the injection motor 35 is connected to the screw shaft of the ball screw 351 via the belt 353 .
- a first movable support 38 is fixed to the nut 352 of the ball screw 351 .
- a screw 32 is rotatably attached to the first movable support 38 .
- the main body of the weighing motor 34 is fixed to the first movable support 38 .
- the first movable support 38 moves according to the operation of the injection motor 35, and the screw 32 and the metering motor 34 move.
- a first movable support 38 is movably mounted on a guide rail 37 fixed to the base 4 .
- the direction toward the mold clamping device 2 can be called the front, and the direction away from the mold clamping device 2 can be called the rear.
- the cylinder 31 advances toward the mold clamping device 2 by receiving the driving force from the moving motor 36 and retreats away from the mold clamping device 2 .
- the output shaft of the moving motor 36 is connected to the screw shaft of the ball screw 361 .
- the second movable support 39 is coupled to the nut 362 of the ball screw 361 via an elastic member (for example, spring) 363 .
- the rear end of the cylinder 31 is fixed to the second movable support 39 .
- the second movable support 39 and the cylinder 31 move according to the operation of the moving motor 36 .
- a second movable support 39 is movably mounted on a guide rail 37 fixed to the base 4 .
- each motor may incorporate a meter such as an encoder. The motor is feedback-controlled based on the output signal of the encoder.
- a backflow prevention ring (not shown) is attached to the tip (front end) of the screw 32 .
- the anti-backflow ring prevents the molten plastic material stored in the storage space 31e from flowing back when the screw 32 is moved toward the nozzle portion 31b in the cylinder 31. As shown in FIG.
- the heater 33 is attached to the outer periphery of the cylinder 31 and generates heat, for example, by feedback-controlled energization.
- the heater 33 is attached to the outer periphery of the cylinder body portion 31a and/or the nozzle portion 31b in any manner.
- the heater 33 gives heat to the cylinder 31, and the pellets fed into the cylinder body 31a through the hopper 31f are melted.
- the screw 32 rotates in the cylinder body 31a according to the rotational force from the weighing motor 34, and the plastic material is fed forward along the helical groove of the screw 32. In this process, the plastic material is gradually melted. be.
- the screw 32 is retracted, and the molten plastic material is stored in the storage space 31e (referred to as "weighing process").
- the number of rotations of the screw 32 is detected using the encoder of the metering motor 34 .
- the injection motor 35 may be driven to apply back pressure to the screw 32 to limit its rapid retraction.
- Back pressure on the screw 32 is detected using, for example, a pressure detector.
- the screw 32 is retracted to the metering completion position, a predetermined amount of molten plastic material is accumulated in the storage space 31e in front of the screw 32, and the metering process is completed.
- the screw 32 moves toward the nozzle portion 31b from the filling start position to the filling completion position in accordance with the driving force from the injection motor 35, and the molten plastic material stored in the storage space 31e flows into the nozzle portion 31b. It is supplied into the mold device 5 through the ejection port (called a “filling step”).
- the position and speed of the screw 32 are detected using an encoder of the injection motor 35, for example.
- V/P switching switching from the filling process to the holding pressure process
- the position at which V/P switching takes place is also called the V/P switching position.
- the set speed of the screw 32 may be changed according to the position of the screw 32, time, and the like.
- the screw 32 When the position of the screw 32 reaches the set position in the filling process, the screw 32 may be temporarily stopped at the set position, and then V/P switching may be performed. Just before the V/P switching, instead of stopping the screw 32, the screw 32 may be slowly advanced or slowly retracted. Also, the screw position detector that detects the position of the screw 32 and the screw speed detector that detects the speed of the screw 32 are not limited to the encoder of the injection motor 35, and other types of detectors can be used.
- the holding pressure of the plastic material in front of the screw 32 is maintained at the set pressure according to the forward movement of the screw 32, and the remaining plastic material is extruded into the mold device 5 (referred to as the "holding pressure process"). be called).
- the shortage of plastic material due to cooling shrinkage in the mold device 5 can be replenished.
- the holding pressure is detected using, for example, a pressure detector.
- the set value of the holding pressure may be changed according to the elapsed time from the start of the holding pressure process.
- the plastic material in the cavity inside the mold device 5 is gradually cooled, and when the holding pressure process is completed, the entrance of the cavity is closed with the solidified plastic material.
- This condition is called a gate seal and prevents backflow of plastic material from the cavity.
- the cooling process is started. In the cooling process solidification of the plastic material of the cavity takes place.
- the metering step of the next molding cycle may be started during the cooling step.
- the injection molding machine 1 has a control panel 7 (see FIG. 1) in which a control system for controlling the mold clamping device 2 and/or the injection device 3 is stored.
- a control system housed in the control panel 7 sequences the mold clamping motor 26 , the ejector motor, the metering motor 34 and the injection motor 35 .
- the control system performs mold closing, mold clamping, and mold opening based on the control of the mold clamping motor 26 .
- the control system performs metering, filling, and holding pressure based on control of metering motor 34 and injection motor 35 .
- the control system can eject the molded product from the movable mold 52 of the mold device 5 based on the control of the ejector motor.
- the control system can position the cylinder 31 to the proper position under control of the movement motor 36 .
- the control system can also control the temperature of the heater 33 and the mold device 5 in addition to the control described above.
- the weighing process, mold closing process, mold clamping process, filling process, holding pressure process, cooling process, mold opening process, and ejecting process are performed in this order.
- the order described here is the order of the start time of each step.
- the filling process, holding pressure process, and cooling process are performed from the start of the mold clamping process to the end of the mold clamping process.
- the end of the mold closing process coincides with the start of the mold opening process.
- a plurality of steps may be performed at the same time.
- the metering step may occur during the cooling step of the previous molding cycle, in which case the mold closing step may occur at the beginning of the molding cycle.
- the filling process may also be initiated during the mold closing process.
- the ejecting process may be initiated during the mold opening process.
- an injection molding machine control section 60 is provided so as to be communicable with the injection molding machine main body 1' including the mold clamping device 2 and the injection device 3 described above.
- the injection molding machine main body 1' performs injection molding based on input of injection molding conditions (that is, a combination of two or more individual setting conditions) to manufacture molded products.
- the injection molding machine controller 60 may be incorporated in the control panel 7 described above, or may be provided separately from the control panel 7 described above.
- the injection molding machine controller 60 can be embodied in a computer.
- At least one CPU Central Processing Unit
- at least one memory hard drive, semiconductor memory
- a program read from the memory is executed by the CPU
- a desired function e.g., prediction model generation unit 65, program modules such as the injection molding condition determination unit 66
- the injection molding machine control unit 60 has a data storage unit 61 , an injection molding condition search support unit 64 , an injection molding condition setting unit 67 , and a buffer unit 68 .
- the data storage unit 61 is a database in which advance data is stored.
- the prior data consists of (i) injection molding condition data, (ii) at least one quality data (data represented by continuous and/or discrete values) associated with each other.
- Injection molding condition data (individual setting conditions) and quality data both take real numbers, but are not necessarily limited to this.
- Quality data may include continuous quality values.
- the quality data may include discrete valued quality indicators.
- injection molding conditions include two or more individual conditions (individual conditions represented by continuous and/or discrete values).
- injection molding condition data X (X 1 -X 4 ) and quality data Y (Y 1 -Y 4 ) can be presented.
- injection molding condition data X and quality data Y other than those shown in Table 1 can also be used. That is, any set condition X 1 -X 4 and any quality Y 1 -Y 4 presented in Table 1 can be excluded.
- Various molding conditions can be adopted as explanatory variables, and similar various qualities can be adopted as objective variables.
- X 1 to X 4 shown in Table 1 are to be understood as non-limiting examples with respect to their individual content and their order.
- the number of setting conditions is also four for easy understanding. In an injection molding machine, the number of setting conditions is generally 10 or more, and the burden of searching for injection molding conditions is enormous.
- the probability of occurrence of a specific phenomenon related to the quality of an injection-molded product is represented by discrete values, where the first real number "1" is assigned to the presence of burrs or sink marks, and the second real number "0" is assigned to the absence of burrs or sink marks. assigned to.
- Discrete quality evaluation is not limited to using two evaluation values, and it is also possible to use three or more evaluation values.
- the injection molding condition search support unit 64 has a prediction model generation unit 65 and an injection molding condition determination unit 66 .
- the search support unit 64 performs SMBO (Sequential Model-based Optimization) (for example, Bayesian optimization) using a probabilistic prediction model with respect to quality represented by continuous values. (2) the value of the objective function (i.e., the first injection Optimizing (maximizing or minimizing) the value that indicates the feasibility of the molding conditions. The injection molding condition when the value of the objective function becomes the optimum value is selected as the next injection molding condition. Additionally or alternatively, the search support unit 64 performs classification predictions such as logistic regression and Gaussian process identification, and in short, (1) for calculating the probability distribution regarding the probability of occurrence of a specific phenomenon related to the quality of the molded product.
- SMBO Simple Model-based Optimization
- the value of the objective function (that is, the value indicating the likelihood of the second injection molding condition), which is a function of the probability of occurrence of two or more probability distributions regarding two or more specific phenomena ( maximize or minimize).
- the injection molding condition that gives the optimum value of the objective function is selected as the next injection molding condition.
- an objective function that is a function of the expected value (or average value) and variance (e.g., standard deviation) of the first probability distribution and is a function of the occurrence probability of one probability distribution related to one specific phenomenon can also be used.
- the prediction model generation unit 65 generates a prediction model that predicts the quality of unknown injection molding conditions based on prior data in which injection molding condition data and molded product quality data are associated. Quality is represented by continuous values and by discrete values. Therefore, different prediction models are generated depending on whether the quality is continuous or discrete. Specifically, as a prediction model, (i) a first probability distribution in which the predicted value and variance of quality continuously change according to changes in injection molding conditions, or (ii) a specified probability distribution according to changes in injection molding conditions A method is derived for calculating a second probability distribution in which the probability of occurrence of a phenomenon varies continuously.
- the prediction model (i) may be called a regression model, and the prediction model (ii) may be called a classification model for distinction.
- the injection molding condition determination unit 66 evaluates one or more first probability distributions regarding quality and/or evaluates two or more second probability distributions regarding two or more specific phenomena, and/or evaluates at least one first probability A next injection molding condition to be set in the injection molding machine is determined based on the evaluation of the distribution and the at least one second probability distribution. Specifically, the injection molding condition determination unit 66 determines parameters (eg, expected value (or average value), variance (eg, standard deviation)) or variables (eg, occurrence probability) related to the first and/or second probability distributions. ), the probability distribution is evaluated using an objective function for calculating a value indicating the likelihood of the injection molding conditions.
- parameters eg, expected value (or average value), variance (eg, standard deviation)
- variables eg, occurrence probability
- the injection molding condition when the value of the objective function becomes the optimum value (maximum or minimum value) is determined as the next injection molding condition to be set in the injection molding machine main body 1'.
- Such cooperation between the predictive model generation unit 65 and the injection molding condition determination unit 66 makes it possible to obtain probabilistically promising (that is, better quality molded products) in light of past injection molding conditions and past quality. It is possible to determine the following injection molding conditions.
- the probabilistic prediction model as described above, it is possible to obtain a prediction distribution of molding quality corresponding to unknown injection molding conditions, and molding conditions that are expected to stochastically exceed the best quality of prior data. can be specified. Therefore, it is useful for searching for injection molding conditions even at a stage where there is little prior data, and is particularly useful for those who have no knowledge or experience in searching for injection molding conditions. It should be noted that the next injection molding condition may be determined regardless of quality or with low correlation at the stage where prior data is scarce.
- the injection molding condition determination unit 66 uses an objective function to calculate a value indicating the likelihood of the injection molding condition from the parameters or variables relating to the first and/or second probability distributions, as described above. Evaluate distributions, but are not limited to this. Evaluation of the probability distribution can also be performed in multiple stages, for example, a range of promising injection molding conditions can be determined, and then injection molding conditions can be specified within that range. It is also possible to obtain a plurality of promising injection molding conditions and select a desired one from among them. Various computational techniques, such as weighting, are available for determining values that indicate the likelihood of injection molding conditions.
- the quality value may contain noise.
- noise For example, there is a possibility that the evaluation results of quality (especially the presence or absence of appearance abnormalities such as burrs and sink marks) will vary between experts and beginners.
- the probabilistic prediction model it is possible to determine probabilistically promising next injection molding conditions in consideration of the influence of such noise. This increases the chances of obtaining a molded part of better quality.
- the prediction model generation unit 65 has a regression model derivation unit 71, a classification model derivation unit 72, and a model storage unit 73.
- the regression model derived by the regression model derivation unit 71 is stored in the model storage unit 73 .
- the classification model derived by the classification model derivation unit 72 is stored in the model storage unit 73 . Both the regression model derivation unit 71 and the classification model derivation unit 72 can be employed, or only one of them can be employed.
- the regression model derivation unit 71 may generate a model for calculating probability distributions based on/according to a Gaussian process (thus using the model to calculate the expected value (or mean) and variance of quality for certain injection molding conditions). is obtained). An average value can also be obtained in addition to or as an alternative to the expected value.
- the unknown quality y for any injection molding condition x is obtained as a Gaussian probability distribution. That is, the expected value ⁇ (x) and variance ⁇ (x) of the predicted value of quality are obtained.
- Equation 1 shows a probability distribution model of molding quality ynew corresponding to unknown injection molding condition xnew .
- D denotes a known (x,y).
- Equation 1 The distribution of the molding quality y new corresponding to the unknown injection molding condition x new is expressed on the right side of Equation 1 as is represented by k * is a vector representation of the relationship between x new and known x 1 to x N using a kernel function.
- k ** is a scalar representation of the relationship between x new and x new using a kernel function.
- K is a matrix representation of the known relationships between x 1 to x N using a kernel function.
- Equation 2 The expected value of the predicted Gaussian distribution can be expressed as shown in Equation 2.
- the variance can be expressed as shown in Equation 3.
- a kernel function (for example, the Gaussian kernel of Equation 4) is used in calculating the prediction distribution.
- ⁇ of the kernel function a point-estimated value that maximizes the marginal likelihood of the Gaussian process can be used, but the present invention is not limited to this. Hyperparameters can also be estimated as random variables.
- FIG. 4 shows an image example of the probability distribution calculated from the model.
- the molding conditions X 10 to X 30 and quality values corresponding to these are included in the preliminary data.
- the solid line in FIG. 4 indicates that the expected value of quality changes continuously as the injection molding conditions change.
- the dashed-dotted line in FIG. 4 schematically shows a Gaussian distribution with the expected value of quality as a vertex.
- the two dotted lines above and below the solid line in FIG. 4 define a credible interval based on a given probability, for example a 95% probability credible interval.
- the quality will be a value within the credible interval with a given probability (eg, 95% probability).
- a good implementation of the Gaussian process will result in a small variance in quality predictions for known injection molding conditions and a high variance in quality predictions for unknown injection molding conditions.
- Using such a probability distribution (specifically, using the expected value (or average value) and variance of quality), it is possible to predict quality for unknown injection molding conditions. More specifically, the molding condition X next is more likely than the molding condition X 10 to obtain a quality value exceeding the molding quality target value, and the molding condition X next can be selected as a promising molding condition.
- the injection molding conditions appear to be one variable in FIG. 4, the injection molding conditions are actually determined by a plurality of setting conditions.
- Injection molding conditions that are worth exploring using an objective function (e.g., to calculate a value that indicates the likelihood of an injection molding condition as a function of the expected value and variance of the predicted quality) (e.g., the objective function Injection molding conditions when the value becomes the optimum value) can be obtained.
- an objective function e.g., to calculate a value that indicates the likelihood of an injection molding condition as a function of the expected value and variance of the predicted quality
- the classification model deriving unit 72 is configured to generate a model for calculating the occurrence probability of a specific phenomenon related to the quality of the molded product from the injection molding conditions (two or more individual setting conditions). Generate such a model based on the instrument.
- logistic regression for example, a logistic function is used to fit the probability of occurrence, thereby obtaining a regression curve as shown in FIG.
- the probability of occurrence of a specific phenomenon continuously changes according to changes in injection molding conditions.
- a regression curve presents an area where the probability of occurrence of a specific phenomenon changes from a low value to a high value, and it is possible to evaluate the probability distribution by focusing on this.
- Injection molding conditions worth searching for using an objective function for example, calculating the sum or logarithmic sum of occurrence probabilities of two or more specific phenomena) molding conditions) can be obtained.
- the upper part of FIG. 5 shows an example of the probability distribution regarding the occurrence of sink marks.
- the upper solid line in FIG. 5 indicates the sink mark occurrence probability.
- the area determined by the molding condition value X 20 and the molding condition value X 30 is the area where there is a transition from the presence of sink marks to the absence of sink marks, and the quality value related to sink marks is the defective product value (indicating that the product is defective). This is the area where the value changes from 100% to 100% (indicating that the product is good).
- the area determined by the molding condition value X20 and the molding condition value X30 is the area where there is a transition from no burrs to burrs. This is the area where the defective product value transitions (indicating that the product is defective).
- FIG. 5 there is an area where the probability of occurrence of a specific phenomenon changes from a low value (eg, 0.2 or less) to a high value (eg, 0.8 or more), and an area where the occurrence probability of another specific phenomenon is high. It may be useful to select injection molding conditions that belong to both areas varying from values (eg, 0.8 or higher) to low values (eg, 0.2 or lower). Quality is not limited to correlated qualities such as burrs and sink marks.
- the injection molding condition determination unit 66 determines parameters (eg, expected value (or average value), variance (eg, standard deviation)) or variables (eg, occurrence probability) related to the first and/or second probability distributions. ) to determine the next injection molding conditions to be set for the injection molding machine main body 1′ using an objective function for calculating a value indicating the likelihood of the injection molding conditions. Specifically, the molding conditions are selected when the value of the objective function becomes the optimum value. Gradient method and MCMC (Metropolis-Hasting) method are available as optimization tools for optimizing (maximizing or minimizing) the value of the objective function.
- parameters eg, expected value (or average value), variance (eg, standard deviation)
- variables eg, occurrence probability
- An objective function relating to an upper confidence bound can be used for the evaluation of the first probability distribution of (i).
- UCB(x) indicates the value of UCB under injection molding condition x
- ⁇ (x) indicates the expected value of quality predicted under injection molding condition x
- ⁇ (x) denotes the standard deviation of the expected quality at injection molding condition x
- k indicates a hyperparameter.
- PI (x) indicates the value of PI under injection molding conditions x
- ⁇ (x) indicates the expected value of quality predicted under injection molding condition x
- ⁇ (x) denotes the standard deviation of the expected quality at injection molding condition x
- y denotes the quality value
- y max indicates the maximum value of quality.
- maximizing UCB(x) means that the sum of the expected quality value and variance (eg, standard deviation) increases.
- a molded product of better quality can be obtained stochastically.
- k which is a hyperparameter
- the search for unknown molding conditions is prioritized.
- Equation 6 similarly to Equation 5, a molded product of better quality can be stochastically obtained by obtaining the injection molding condition x when PI(x) is maximized. Determining the injection molding condition x in this way is equivalent to evaluating the probability distribution based on the objective function.
- Equation 7 the product (P all (x)) of the individual occurrence probabilities (P n (x)) of N (N is a natural number of 2 or more) quality under the injection molding condition x is calculated. .
- Expression 8 the logarithmic value of the product of occurrence probabilities in Expression 7 is calculated, and can be understood in the same way as Expression 7.
- Equation 7 maximization of P all (x) means that the probability of occurrence (P n (x)) is increased as an overall trend in N qualities.
- the expected value and the variance are incorporated into the objective function (thus, the value of the objective function (the injection The value indicating the likelihood of the molding condition) becomes a function of the expected value and the variance), and in the case of the second probability distribution of (ii), the probability of occurrence is incorporated into the objective function (therefore, the value of the objective function (injection molding The plausibility value of the condition) is a function of the probability of occurrence).
- the injection molding condition x when the value of the objective function is optimized eg, maximized or minimized
- a probable injection molding condition x that results in a molded product of better quality is determined.
- the injection molding condition x when the value of the objective function is maximized is calculated.
- the injection molding condition x when the value of the objective function is minimized is calculated.
- An optimization tool is used to optimize the value of the objective function. For example, gradient method, MCMC (Metropolis-Hasting) method can be used.
- the objective function can specify the search range. If there is a target value for some quality, the target value can be defined in the objective function. When optimizing for multiple qualities, the importance of these qualities can be set. It is also possible to maximize or minimize quality for which there is no target value.
- the evaluation of (i) the first probability distribution and the evaluation of (ii) the second probability distribution can also be performed using a common objective function. That is, the objective function for the first probability distribution and the objective function for the second probability distribution are combined to form a common objective function.
- this objective function is the function of UCB(x) in Equation 5 and P all (x) in Equation 7.
- Equation 6 can be used in place of Equation 5
- Equation 8 can be used in place of Equation 7 for such integration of objective functions.
- the injection molding condition determination unit 66 has an objective function storage unit 74 and an arithmetic execution unit 75, as shown in FIG.
- An objective function is stored in the objective function storage unit 74 .
- the calculation execution unit 75 incorporates parameters or variables relating to the probability distribution calculated from the prediction model stored in the model storage unit 73 into the objective function, and uses an optimization tool to obtain the value of the objective function (probability of the injection molding conditions). Find the injection molding condition x when the value shown) becomes the optimum value. In this way, the following injection molding conditions are determined.
- the molding conditions under which the value of the objective function becomes the optimum value are not necessarily the molding conditions under which a molded product of better quality is expected to be obtained. For example, at a stage where there is little prior data for deriving a prediction model, useful molding conditions are selected to improve prediction accuracy.
- the injection molding condition x next that is most likely to exceed the target quality is specified and can be selected.
- the injection molding condition x next that is most likely to exceed the target quality is specified and can be selected.
- the upper bound of the 95% credible interval is chosen.
- the selected conditions are balanced in terms of both distance from known molding conditions and quality values.
- injection molding conditions are chosen that are likely to exceed the known maximum for quality.
- the injection molding condition determining unit 66 can select molding conditions with relatively large dispersion of predicted values of quality as the next injection molding conditions. By selecting molding conditions with a large quality variance, it is possible to suppress the selection of injection molding conditions that are spatially or in coordinate space close to known injection molding conditions, and to enable efficient search for injection molding conditions. .
- the injection molding condition x that maximizes the objective function is defined as an area where the probability of occurrence of a specific phenomenon changes from a low value (e.g., 0.2 or less) to a high value (e.g., 0.8 or more) and another specific phenomenon. It can be an injection molding condition belonging to both areas where the probability of occurrence of changes from a high value (eg, 0.8 or more) to a low value (eg, 0.2 or less).
- the injection molding condition setting unit 67 instructs the injection molding machine body 1' to operate based on the received search candidate injection molding conditions.
- the injection molding machine main body 1 ′ operates based on the injection molding conditions designated by the injection molding condition setting section 67 to manufacture molded products.
- Injection molding may be performed after confirmation by the user.
- the user inspects the part and enters the quality data into the buffer section.
- the injection molding condition data and the quality data are stored in the buffer section 68 in association with each other.
- the user stores the data stored in the buffer section 68 in the data storage section 61 using input means (not shown). New data is added in this way.
- the prediction model generation unit 65 can generate a new prediction model using the updated database with new data added.
- the user sets an objective function (S1).
- the user operates the injection molding machine 1 to obtain and register prior information in which the injection molding condition data and the quality data are associated with each other (S2). For example, prior data as shown in Table 1 are registered in the database.
- the prediction model generator 65 generates a prediction model based on the prior data prepared in S2 (S3).
- the predictive model may be for calculating either or both of the first or second probability distributions as described above.
- the injection molding condition determination unit 66 determines injection molding conditions for the next search candidate using an objective function that is a function of parameters or variables of the probability distribution (S4).
- the injection molding condition x under which the value of the objective function becomes the optimum value is selected as the next injection molding condition.
- the injection molding conditions determined in this manner are set by the injection molding condition setting unit 67 as operating conditions for the injection molding machine main body 1' (S5), and the injection molding machine main body 1' operates under these conditions (S6).
- the quality of the molded product manufactured in this manner is evaluated by a user or a quality judgment device (S7).
- the injection molding condition data and the quality data are added as prior data associated with each other (S8).
- the prediction model is derived again, the loop of S3 to S8 is repeated, and the prediction model continues to be updated.
- Injection molding machine 61 Data storage unit 65: Prediction model generation unit 66: Injection molding condition determination unit
Landscapes
- Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Mechanical Engineering (AREA)
- Injection Moulding Of Plastics Or The Like (AREA)
Abstract
Description
射出成形条件データと成形品の品質データが関連付けられた事前データに基づいて未知の射出成形条件に関する品質を予測する予測モデルを生成する予測モデル生成部にして、予測モデルは、(i)射出成形条件の変化に応じて品質の予測値及び分散が連続的に変化する第1確率分布を算出するための予測モデル、及び/又は(ii)射出成形条件の変化に応じて品質に関する特定現象の発生確率が連続的に変化する第2確率分布を算出するための予測モデルを含む、予測モデル生成部と、
1以上の品質に関する第1確率分布の評価、及び/又は、2以上の特定現象に関する2以上の第2確率分布の評価、及び/又は、少なくとも一つの第1確率分布と少なくとも一つの第2確率分布の評価に基づいて、射出成形機に設定されるべき次の射出成形条件を決定する射出成形条件決定部を含む。
射出成形条件データと成形品の品質データが関連付けられた事前データに基づいて未知の射出成形条件に関する品質を予測する予測モデルを生成し、予測モデルは、(i)射出成形条件の変化に応じて品質の予測値及び分散が連続的に変化する第1確率分布を算出するための予測モデル、及び/又は(ii)射出成形条件の変化に応じて品質に関する特定現象の発生確率が連続的に変化する第2確率分布を算出するための予測モデルを含み、
1以上の品質に関する第1確率分布の評価、及び/又は、2以上の特定現象に関する2以上の第2確率分布の評価に基づいて、射出成形機に設定されるべき次の射出成形条件を決定する。
により表される。k*は、xnewと既知のx1~xNの関係性をカーネル関数によりベクトル表現したものである。k**は、xnewとxnewの関係性をカーネル関数によりスカラー表現したものである。Kは、既知のx1~xNの関係性をカーネル関数により行列表現したものである。
UCB(x)は、射出成形条件xにおけるUCBの値を示し、
μ(x)は、射出成形条件xにおいて予測される品質の期待値を示し、
σ(x)は、射出成形条件xにおける予測される品質の標準偏差を示し、
kは、ハイパーパラメータを示す。
PI(x)は、射出成形条件xにおけるPIの値を示し、
μ(x)は、射出成形条件xにおいて予測される品質の期待値を示し、
σ(x)は、射出成形条件xにおける予測される品質の標準偏差を示し、
yは、品質の値を示し、
ymaxは、品質の最大値を示す。
61 :データ記憶部
65 :予測モデル生成部
66 :射出成形条件決定部
Claims (16)
- 射出成形条件の入力に基づいて射出成形を実行して成形品を製造する射出成形機に関する射出成形条件の探索支援装置であって、
射出成形条件データと成形品の品質データが関連付けられた事前データに基づいて未知の射出成形条件に関する品質を予測する予測モデルを生成する予測モデル生成部にして、前記予測モデルは、(i)射出成形条件の変化に応じて品質の予測値及び分散が連続的に変化する第1確率分布を算出するための予測モデル、及び/又は(ii)射出成形条件の変化に応じて品質に関する特定現象の発生確率が連続的に変化する第2確率分布を算出するための予測モデルを含む、予測モデル生成部と、
1以上の品質に関する前記第1確率分布の評価、及び/又は、2以上の特定現象に関する2以上の前記第2確率分布の評価、及び/又は、少なくとも一つの前記第1確率分布と少なくとも一つの前記第2確率分布の評価に基づいて、前記射出成形機に設定されるべき次の射出成形条件を決定する射出成形条件決定部を備える、射出成形条件の探索支援装置。 - 前記射出成形条件決定部は、前記第1及び/又は第2確率分布に関するパラメータ及び/又は変数から射出成形条件の有望さを示す値を算出する目的関数を利用して次の射出成形条件を決定することを特徴とする請求項1に記載の射出成形条件の探索支援装置。
- 前記パラメータは、前記第1確率分布における前記品質の予測値及び分散を含むことを特徴とする請求項2に記載の射出成形条件の探索支援装置。
- 前記変数は、前記第2確率分布における前記特定現象の発生確率を含むことを特徴とする請求項2又は3に記載の射出成形条件の探索支援装置。
- 前記射出成形条件の有望さを示す値は、前記1以上の品質に関する前記第1確率分布における前記品質の予測値及び分散の関数であり、かつ前記2以上の特定現象に関する2以上の前記第2確率分布における前記発生確率の関数であることを特徴とする請求項2乃至4のいずれか一項に記載の射出成形条件の探索支援装置。
- 前記目的関数は、上限信頼区間(UCB(Upper Confidence Bound))又は品質に関する既存最大値を更新する確率(PI(Probability of Improvement))に関することを特徴とする請求項2乃至5のいずれか一項に記載の射出成形条件の探索支援装置。
- 前記目的関数は、前記2以上の特定現象に関する2以上の前記第2確率分布における前記発生確率の積又はこの対数値に関することを特徴とする請求項2乃至6のいずれか一項に記載の射出成形条件の探索支援装置。
- 前記予測モデル生成部は、ガウス過程に基づいて前記第1確率分布を生成することを特徴とする請求項1乃至7のいずれか一項に記載の射出成形条件の探索支援装置。
- 前記射出成形条件決定部は、上限信頼区間(UCB(Upper Confidence Bound))又は品質に関する既存最大値を更新する確率(PI(Probability of Improvement))に基づいて次の射出成形条件を決定するように構成されることを特徴とする請求項8に記載の射出成形条件の探索支援装置。
- 前記射出成形条件決定部は、品質の予測値の分散が相対的に大きい成形条件を次の射出成形条件として選択するように構成されることを特徴とする請求項8又は9に記載の射出成形条件の探索支援装置。
- 前記予測モデル生成部は、ロジスティック回帰又はガウス過程識別器に基づいて、射出成形条件から特定現象の発生確率を算出する分類モデルを生成するように構成されることを特徴とする請求項1乃至10のいずれか一項に記載の射出成形条件の探索支援装置。
- 前記射出成形条件決定部は、前記2以上の特定現象に関する2以上の前記第2確率分布における前記発生確率の積又はこの対数値に基づいて次の射出成形条件を決定するように構成されることを特徴とする請求項11に記載の射出成形条件の探索支援装置。
- 前記射出成形条件決定部は、ある特定現象の発生確率が0.2以下の値から0.8以上の値に変化する区域と、別の特定現象の発生確率が0.8以上の値から0.2以下の値に変化する区域の両方に属する射出成形条件を前記次の射出成形条件として選択するように構成されることを特徴とする請求項11又は12に記載の射出成形条件の探索支援装置。
- 射出成形条件の入力に基づいて射出成形を実行して成形品を製造する射出成形機に関する射出成形条件の探索支援方法であって、
射出成形条件データと成形品の品質データが関連付けられた事前データに基づいて未知の射出成形条件に関する品質を予測する予測モデルを生成し、前記予測モデルは、(i)射出成形条件の変化に応じて品質の予測値及び分散が連続的に変化する第1確率分布を算出するための予測モデル、及び/又は(ii)射出成形条件の変化に応じて品質に関する特定現象の発生確率が連続的に変化する第2確率分布を算出するための予測モデルを含み、
1以上の品質に関する前記第1確率分布の評価、及び/又は、2以上の特定現象に関する2以上の第2確率分布の評価に基づいて、前記射出成形機に設定されるべき次の射出成形条件を決定する、射出成形条件の探索支援方法。 - 射出成形条件データと成形品の品質データが関連付けられた事前データに基づいて未知の射出成形条件に関する品質を予測する予測モデルを生成し、前記予測モデルは、(i)射出成形条件の変化に応じて品質の予測値及び分散が連続的に変化する第1確率分布を算出するための予測モデル、及び/又は(ii)射出成形条件の変化に応じて品質に関する特定現象の発生確率が連続的に変化する第2確率分布を算出するための予測モデルを含み、
1以上の品質に関する前記第1確率分布の評価、及び/又は、2以上の特定現象に関する2以上の第2確率分布の評価に基づいて、射出成形機に設定されるべき次の射出成形条件を決定する処理をコンピュータに実行させるためのプログラム。 - 請求項15に記載のプログラムが記憶された非一時的な記録媒体。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE112022001901.8T DE112022001901T5 (de) | 2021-03-31 | 2022-03-25 | Suchhilfsvorrichtung für spritzgiessbedingung und verfahren dafür, programm und speichermedium |
JP2023511204A JPWO2022210425A1 (ja) | 2021-03-31 | 2022-03-25 | |
CN202280008207.6A CN116600964A (zh) | 2021-03-31 | 2022-03-25 | 注射成型条件的搜索支援装置及其方法、程序及记录介质 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021062347 | 2021-03-31 | ||
JP2021-062347 | 2021-03-31 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022210425A1 true WO2022210425A1 (ja) | 2022-10-06 |
Family
ID=83456251
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/014669 WO2022210425A1 (ja) | 2021-03-31 | 2022-03-25 | 射出成形条件の探索支援装置及びその方法、プログラム及び記録媒体 |
Country Status (4)
Country | Link |
---|---|
JP (1) | JPWO2022210425A1 (ja) |
CN (1) | CN116600964A (ja) |
DE (1) | DE112022001901T5 (ja) |
WO (1) | WO2022210425A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023145549A1 (ja) * | 2022-01-25 | 2023-08-03 | 三菱電機株式会社 | 射出成形方法、成形条件導出装置、およびコンピュータ読み込み可能な記憶媒体 |
US11868900B1 (en) | 2023-02-22 | 2024-01-09 | Unlearn.AI, Inc. | Systems and methods for training predictive models that ignore missing features |
US12008478B2 (en) | 2019-10-18 | 2024-06-11 | Unlearn.AI, Inc. | Systems and methods for training generative models using summary statistics and other constraints |
US12020789B1 (en) * | 2023-02-17 | 2024-06-25 | Unlearn.AI, Inc. | Systems and methods enabling baseline prediction correction |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190171776A1 (en) * | 2017-12-01 | 2019-06-06 | Industrial Technology Research Institute | Methods, devices and non-transitory computer-readable medium for parameter optimization |
WO2020040011A1 (ja) * | 2018-08-24 | 2020-02-27 | 日本電信電話株式会社 | Osnrスペクトル推定装置、osnrスペクトル推定方法およびプログラム |
JP2020075385A (ja) * | 2018-11-06 | 2020-05-21 | 株式会社東芝 | 製品状態推定装置 |
CN112101630A (zh) * | 2020-08-19 | 2020-12-18 | 江苏师范大学 | 一种薄壁塑件注射成型工艺参数多目标优化方法 |
DE102019220453A1 (de) * | 2019-12-20 | 2021-06-24 | Robert Bosch Gesellschaft mit beschränkter Haftung | Verfahren und Vorrichtung zum Erstellen einer Sicherheitsfunktion für das Vermessen eines Fertigungssystems |
WO2021201187A1 (ja) * | 2020-03-31 | 2021-10-07 | 旭化成株式会社 | 装置、方法、およびプログラム |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7159758B2 (ja) | 2018-09-27 | 2022-10-25 | 株式会社ジェイテクト | 成形条件決定支援装置および射出成形機 |
-
2022
- 2022-03-25 DE DE112022001901.8T patent/DE112022001901T5/de active Pending
- 2022-03-25 WO PCT/JP2022/014669 patent/WO2022210425A1/ja active Application Filing
- 2022-03-25 CN CN202280008207.6A patent/CN116600964A/zh active Pending
- 2022-03-25 JP JP2023511204A patent/JPWO2022210425A1/ja active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190171776A1 (en) * | 2017-12-01 | 2019-06-06 | Industrial Technology Research Institute | Methods, devices and non-transitory computer-readable medium for parameter optimization |
WO2020040011A1 (ja) * | 2018-08-24 | 2020-02-27 | 日本電信電話株式会社 | Osnrスペクトル推定装置、osnrスペクトル推定方法およびプログラム |
JP2020075385A (ja) * | 2018-11-06 | 2020-05-21 | 株式会社東芝 | 製品状態推定装置 |
DE102019220453A1 (de) * | 2019-12-20 | 2021-06-24 | Robert Bosch Gesellschaft mit beschränkter Haftung | Verfahren und Vorrichtung zum Erstellen einer Sicherheitsfunktion für das Vermessen eines Fertigungssystems |
WO2021201187A1 (ja) * | 2020-03-31 | 2021-10-07 | 旭化成株式会社 | 装置、方法、およびプログラム |
CN112101630A (zh) * | 2020-08-19 | 2020-12-18 | 江苏师范大学 | 一种薄壁塑件注射成型工艺参数多目标优化方法 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US12008478B2 (en) | 2019-10-18 | 2024-06-11 | Unlearn.AI, Inc. | Systems and methods for training generative models using summary statistics and other constraints |
WO2023145549A1 (ja) * | 2022-01-25 | 2023-08-03 | 三菱電機株式会社 | 射出成形方法、成形条件導出装置、およびコンピュータ読み込み可能な記憶媒体 |
US12020789B1 (en) * | 2023-02-17 | 2024-06-25 | Unlearn.AI, Inc. | Systems and methods enabling baseline prediction correction |
US11868900B1 (en) | 2023-02-22 | 2024-01-09 | Unlearn.AI, Inc. | Systems and methods for training predictive models that ignore missing features |
US11966850B1 (en) | 2023-02-22 | 2024-04-23 | Unlearn.AI, Inc. | Systems and methods for training predictive models that ignore missing features |
Also Published As
Publication number | Publication date |
---|---|
CN116600964A (zh) | 2023-08-15 |
JPWO2022210425A1 (ja) | 2022-10-06 |
DE112022001901T5 (de) | 2024-01-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022210425A1 (ja) | 射出成形条件の探索支援装置及びその方法、プログラム及び記録媒体 | |
JP6529183B2 (ja) | 射出成形機の成形最適化方法 | |
Chen et al. | A review of current developments in process and quality control for injection molding | |
US20200101649A1 (en) | Device for assisting molding condition determination and injection molding apparatus | |
US20080054527A1 (en) | Injection molding apparatus and method of using the same | |
US11537945B2 (en) | Machine learning device and machine learning method | |
US7840306B2 (en) | Molding-system set-up based on molded-part attribute | |
JP6514622B2 (ja) | 射出成形機の成形支援方法 | |
CN111611715B (zh) | 一种注射成形工艺参数无模型优化方法 | |
JP6591260B2 (ja) | 射出成形用情報管理装置、および射出成形機 | |
Huang et al. | Cavity pressure‐based holding pressure adjustment for enhancing the consistency of injection molding quality | |
JP7342450B2 (ja) | 品質予測システム及び成形機 | |
US20200206998A1 (en) | Quality prediction system and molding machine | |
JP7293962B2 (ja) | 成形品の品質異常予測システム | |
US20230241826A1 (en) | Method and device for reducing the amount of reworking required on mold cavities prior to their use in series production | |
WO2022210424A1 (ja) | 品質予測装置及びその方法、プログラム並びに記録媒体 | |
JP2022057084A (ja) | 射出成形機 | |
US20230093303A1 (en) | Method for adapting a component description of a workpiece to be produced with amorphous properties | |
JP2023017386A (ja) | 成形条件調整方法、コンピュータプログラム、成形条件調整装置及び射出成形機 | |
US20220297362A1 (en) | Artificial intelligence-based injection molding system and method for generating molding condition in injection molding system | |
JP7483436B2 (ja) | 射出成形機及びステータス決定装置 | |
WO2023032375A1 (ja) | 成形条件パラメータ調整方法、コンピュータプログラム、成形条件パラメータ調整装置及び成形機 | |
JPH05329873A (ja) | ロータリー成形システム | |
Kazmer | Precision Process Control of Injection Molding | |
JP2009006661A (ja) | 成形方法 |
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: 22780638 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2023511204 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280008207.6 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 112022001901 Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 22780638 Country of ref document: EP Kind code of ref document: A1 |