JP7638816B2 - MOLDING CONDITION ADJUSTMENT METHOD, COMPUTER PROGRAM, MOLDING CONDITION ADJUSTMENT DEVICE, AND INJECTION MOLDING MACHINE - Google Patents

Description

The present invention relates to a molding condition adjustment method, a computer program (program product), a molding condition adjustment device, and an injection molding machine.

There is an injection molding machine system that can appropriately adjust the molding conditions of an injection molding machine through reinforcement learning (for example, Patent Document 1).

JP 2019-166702 A

However, Patent Document 1 does not disclose any specific technology for adjusting the molding conditions of an injection molding machine that uses a mold with a so-called valve gate, that is, a mold equipped with an on-off valve that opens and closes the flow path of the molten molding material that leads to the cavity.

In a mold with a valve gate, if the delay in operation becomes significant due to dirt on the sliding parts of the valve gate, injection occurs when the valve gate's opening and closing valve is not fully open, resulting in molding defects. This poses the problem that the operator must periodically check the opening status of the opening and closing valve and manually adjust the opening timing of the opening and closing valve as necessary, or set a margin to allow for future delays in operation due to dirt on the sliding parts, etc.

The objective of this disclosure is to provide a molding condition adjustment method, computer program, molding condition adjustment device, and injection molding machine that can automatically adjust molding conditions so that the opening timing of the open/close valve matches the injection timing of the molten molding material in an injection molding machine that uses a mold with a so-called valve gate.

The molding condition adjustment method according to this embodiment is a molding condition adjustment method for adjusting the molding conditions of an injection molding machine that uses a mold having a valve gate, and acquires injection pressure waveform data that indicates the time change in injection pressure, and adjusts the molding conditions based on the acquired injection pressure waveform data so that the opening timing of the opening/closing valve and the injection timing of the molten molding material match.

The computer program (program product) according to this embodiment is a computer program for causing a computer to execute a process for adjusting molding conditions of an injection molding machine that uses a mold having a valve gate, and causes the computer to execute a process for acquiring injection pressure waveform data that indicates the change in injection pressure over time, and adjusting the molding conditions based on the acquired injection pressure waveform data so that the opening timing of the opening/closing valve and the injection timing of the molten molding material match.

The molding condition adjustment device according to this embodiment is a molding condition adjustment device that adjusts the molding conditions of an injection molding machine that uses a mold having a valve gate, acquires injection pressure waveform data that indicates the time change in injection pressure, and adjusts the molding conditions based on the acquired injection pressure waveform data so that the opening timing of the opening/closing valve and the injection timing of the molten molding material match.

The injection molding machine according to this embodiment is an injection molding machine that uses a mold having a valve gate and is equipped with the molding condition adjustment device described above.

According to the present invention, in an injection molding machine that uses a mold with a so-called valve gate, the molding conditions can be automatically adjusted so that the opening timing of the opening/closing valve matches the injection timing of the molten molding material.

FIG. 1 is a schematic diagram illustrating an example of the configuration of an injection molding machine according to a first embodiment. 1 is a block diagram showing a configuration example of a molding condition adjustment device according to a first embodiment; FIG. 2 is a conceptual diagram showing a configuration of a table according to the first embodiment. 11 is an explanatory diagram showing an example of an injection pressure waveform and difference information. FIG. 4 is a flowchart showing a processing procedure of a processor according to the first embodiment. FIG. 2 is an explanatory diagram showing a molding condition adjustment method according to the first embodiment. FIG. 11 is a conceptual diagram showing a configuration of a table according to the second embodiment. 10 is a flowchart showing a processing procedure of a processor according to a second embodiment. FIG. 11 is an explanatory diagram showing a molding condition adjustment method according to a second embodiment. FIG. 11 is a block diagram showing a configuration example of a molding condition adjustment device according to a third embodiment. FIG. 11 is a functional block diagram of an injection molding machine according to a third embodiment.

Specific examples of the molding condition adjustment method, computer program (program product), molding condition adjustment device, and injection molding machine according to the embodiments of the present invention will be described below with reference to the drawings. At least some of the embodiments described below may be combined in any manner. Note that the present invention is not limited to these examples, but is indicated by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

Figure 1 is a schematic diagram illustrating an example of the configuration of an injection molding machine 1 according to embodiment 1. The injection molding machine 1 according to embodiment 1 includes a clamping device 2 that clamps a mold 21, an injection device 3 that melts and injects molding material, and a control device 4 that has a molding condition adjustment device 101. The central figure is an enlarged view of the mold 21 portion surrounded by a dashed ellipse.

The mold clamping device 2 includes a fixed platen 22 fixed on the bed 20, a mold clamping housing 23 slidably provided on the bed 20, and a movable platen 24 that similarly slides on the bed 20. The fixed platen 22 and the mold clamping housing 23 are connected by a plurality of, for example, four tie bars 25, 25, .... The movable platen 24 is configured to be slidable between the fixed platen 22 and the mold clamping housing 23. A mold clamping mechanism 26 is provided between the mold clamping housing 23 and the movable platen 24. The mold clamping mechanism 26 is configured, for example, as a toggle mechanism. The mold clamping mechanism 26 may be configured as a direct pressure type mold clamping mechanism, that is, a mold clamping cylinder. The fixed platen 22 and the movable platen 24 are provided with a fixed mold 27 and a movable mold 28, respectively, and the mold 21 is opened and closed when the mold clamping mechanism 26 is driven.

The mold 21 includes a fixed mold 27 and a movable mold 28 opposed to the fixed mold 27 .
The fixed mold 27 includes a fixed-side mold plate 27a and a fixed-side attachment plate 27b, and the fixed-side attachment plate 27b is fixed to the fixed platen 22. A sprue bushing 27c that defines a sprue is formed in the fixed-side mold plate 27a and the fixed-side attachment plate 27b.
On the other hand, the movable mold 28 is clamped to the fixed mold 27 to define a cavity 27d, and includes a movable side mold plate 28a, a movable side mounting plate 28b, a receiving plate 28c, a spacer block 28d, an ejector plate 28e, and an ejector pin 28f.

The fixed platen 22 has a flow path 22a that runs from one side to which the fixed side mounting plate 27b is fixed to the other side, and an opening (valve gate) on one end of the flow path 22a is connected to the sprue bushing 27c, and an opening on the other end is formed on the other side. Around the opening formed on the other side, an annular contact part 22b is provided to which the nozzle 31a of the injection device 3 is in close contact. The molten molding material injected from the nozzle 31a of the injection device 3 is supplied to the cavity 27d of the mold 21 through the flow path 22a and the sprue. A heating device 22c is provided around the flow path 22a to keep the molten molding material flowing through the flow path 22a warm.

The fixed platen 22 also includes an opening/closing valve 29 that opens and closes the flow path 22a. As the opening/closing valve 29, a valve having a well-known structure generally referred to as a valve gate may be appropriately adopted. For example, the opening/closing valve 29 is a mechanical needle valve system that is provided so as to be movable forward and backward with respect to the flow path 22a and includes a needle 29a that opens and closes the flow path 22a, and an air cylinder 29b that drives the needle 29a. The air cylinder 29b can open and close the flow path 22a by moving the needle 29a forward and backward with respect to the flow path 22a. The operation of the opening/closing valve 29 is controlled by the control device 4.

The injection device 3 is provided on a base 30. The injection device 3 includes a heating cylinder 31 having a nozzle 31a at its tip, and a screw 32 arranged within the heating cylinder 31 so as to be rotatable in the circumferential and axial directions. The screw 32 is driven in the rotational and axial directions by a drive mechanism 33. The drive mechanism 33 is composed of a rotary motor that drives the screw 32 in the rotational direction, and a motor that drives the screw 32 in the axial direction. Note that the drive mechanism 33 shown in FIG. 1 is covered with a cover, so the internal configuration is not shown.

A load cell 33a that detects the injection pressure is provided inside the drive mechanism 33, and the detection value data of the injection pressure detected by the load cell 33a is output to the control device 4. The load cell 33a is an example of a sensor that detects the injection pressure. The injection pressure may be directly detected by a pressure sensor provided in the heating cylinder 31.

A hopper 34 into which molding material is poured is provided near the rear end of the heating cylinder 31. The injection molding machine 1 also includes a nozzle touch device 35 that moves the injection device 3 in the front-rear direction (left-right direction in FIG. 1). When the nozzle touch device 35 is driven, the injection device 3 moves forward and the nozzle 31a of the heating cylinder 31 touches the contact portion 22b.

The control device 4 controls the operation of the injection device 3, the opening/closing valve 29, and the mold clamping device 2. The control device 4 according to the present embodiment 1 includes a molding condition adjustment device 101. The molding condition adjustment device 101 is a device that adjusts parameters related to the molding conditions of the injection molding machine 1. In particular, the molding condition adjustment device 101 according to the present embodiment 1 adjusts parameters required to match the opening timing of the opening/closing valve 29 with the injection timing of the molten molding material. The molding condition adjustment device 101 according to the present embodiment 1 performs control to match the opening timing of the opening/closing valve 29 with the injection timing by adjusting the open waiting time from the start of the molding process cycle or the start of mold closing to the start of the opening drive of the opening/closing valve 29.

Note that "matching" does not necessarily mean that the opening timing of the on-off valve 29 and the injection timing of the molten molding material are perfectly synchronized, but rather that the timing is matched to the extent that no abnormalities occur in the injection molding.

In addition to the above-mentioned open waiting time, parameters that determine molding conditions such as injection start time, resin temperature in the mold, nozzle temperature, cylinder temperature, hopper temperature, mold clamping force, injection speed, injection acceleration, injection peak pressure, injection stroke, etc. are set in the injection molding machine 1. Parameters that determine molding conditions such as cylinder tip resin pressure, backflow prevention ring seating state, holding pressure switching pressure, holding pressure switching speed, holding pressure switching position, holding pressure completion position, cushion position, metering back pressure, metering torque, etc. are set in the injection molding machine 1. Parameters that determine molding conditions such as metering completion position, screw retreat speed, cycle time, mold closing time, injection time, holding pressure time, metering time, mold opening time, etc. are set in the injection molding machine 1. Then, the injection molding machine 1 with these parameters set operates according to the parameters. The optimal parameters differ depending on the environment of the injection molding machine 1 and the molded product.

2 is a block diagram showing an example of the configuration of the molding condition adjustment device 101 according to embodiment 1. The molding condition adjustment device 101 is a computer, and includes a processor 111, a storage unit 112, an operation unit 113, and the like as a hardware configuration.
The molding condition adjustment device 101 may be a server device connected to a network. The molding condition adjustment device 101 may be configured to perform distributed processing using multiple computers, may be realized by multiple virtual machines installed in one server, or may be realized by using a cloud server.

The processor 111 includes a central processing unit (CPU), a multi-core CPU, a graphics processing unit (GPU), a general-purpose computing on graphics processing units (GPGPU), a tensor processing unit (TPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a neural processing unit (NPU), and other arithmetic circuits, a read only memory (ROM), a random access memory (RAM), and other internal storage devices, an I/O terminal, and the like. The processor 111 functions as an acquisition unit 114 and a control unit 115 by executing a computer program (program product) 12a stored in a storage unit 112 described below. Each functional unit of the molding condition adjustment device 101 may be realized by software, or may be realized in part or in whole by hardware.

The storage unit 112 is a non-volatile memory such as a hard disk, an EEPROM (Electrically Erasable Programmable ROM), or a flash memory. The storage unit 112 stores a computer program 112a for causing a computer to execute an adjustment process for the open wait time. The storage unit 112 also stores a table 112b for adjusting the open wait time.

Figure 3 is a conceptual diagram showing the configuration of table 112b according to embodiment 1. Table 112b is information that associates injection pressure waveform difference information with the amount of change in open waiting time. The difference information is difference information that indicates the difference between injection pressure waveform data that can be acquired by acquisition unit 114 and predetermined injection pressure waveform data. The acquired injection pressure waveform data is data that indicates the time change of injection pressure. The predetermined injection pressure waveform data is injection pressure waveform data obtained when molding is performed in an ideal state when the opening timing of opening/closing valve 29 and the injection timing of molten molding material match.

Figure 4 is an explanatory diagram showing an example of an injection pressure waveform and difference information. The horizontal axis of the graph shown in Figure 4 indicates the stroke of the screw of the injection device 3, and the vertical axis indicates the injection pressure. In the molding process, the screw moves in the axial direction, and the stroke corresponds to the time in the molding process. The curve shown by the dashed line is the above-mentioned predetermined injection pressure waveform, and the curve shown by the solid line is the injection pressure waveform that can be acquired by the acquisition unit 114. The solid line shown in Figure 4 is the injection pressure waveform obtained when the opening and closing valve 29 opens with a delay from the injection timing of the molten molding material. Since the opening and closing valve 29 is not open at the time injection starts, the injection pressure increases and a peak occurs in the injection pressure waveform. It is possible to estimate the state of the opening and closing valve 29 from the difference between the acquired current injection pressure waveform and the injection pressure waveform in an ideal state.

The difference information is, for example, the area surrounded by the injection pressure waveform shown by the solid line and the specified injection pressure waveform shown by the dashed line. In other words, it can be said to be the absolute value of the difference between the integral value over the stroke of the injection pressure waveform shown by the solid line and the integral value over the stroke of the specified injection pressure waveform. The difference information may be the area obtained by subtracting the integral value of the waveform related to the specified injection pressure waveform from the integral value of the waveform related to the acquired injection pressure waveform data. This area may also be a negative value. In this case, it is possible to deal with not only malfunctions when the acquired injection pressure waveform is higher than the specified injection pressure waveform, but also malfunctions when the acquired injection pressure waveform is lower than the specified injection pressure waveform.

The difference in area is an example of difference information, and is not particularly limited as long as it is information that expresses the difference between each injection pressure waveform. For example, the difference between a coefficient that determines the approximation curve of the waveform indicated by the acquired injection pressure waveform data and a coefficient that determines the approximation curve of the waveform indicated by the specified injection pressure waveform data may be obtained as the difference information. The injection pressure waveform may be approximated using a trigonometric function, an exponential function, a logarithmic function, a polynomial, etc.

The computer program 112a according to the first embodiment may be recorded in a computer-readable manner on the recording medium 5. The storage unit 112 stores the computer program 112a read from the recording medium 5 by a reading device. The recording medium 5 is a semiconductor memory such as a flash memory. The recording medium 5 may also be an optical disk such as a CD (Compact Disc)-ROM, a DVD (Digital Versatile Disc)-ROM, or a BD (Blu-ray (registered trademark) Disc). The recording medium 5 may also be a magnetic disk such as a flexible disk or a hard disk, a magnetic optical disk, or the like. The computer program 112a according to the first embodiment may also be downloaded from an external server connected to a communication network and stored in the storage unit 112.

The operation unit 113 is an input device such as a touch panel, soft keys, hard keys, a keyboard, or a mouse.

The acquisition unit 114 acquires, as injection pressure waveform data, the time-series detection data measured and output by the load cell 33a when molding is performed by the injection molding machine 1. The acquisition unit 114 outputs the acquired injection pressure waveform data to the control unit 115.

Based on the injection pressure waveform data acquired by the acquisition unit 114, the control unit 115 adjusts the molding conditions so that the opening timing of the opening/closing valve 29 matches the injection timing of the molten molding material.

The molding process cycle is outlined below, and the molding condition adjustment device 101 adjusts the open wait time of the on-off valve 29 during the repeated molding process cycle. During injection molding, the well-known mold closing process, mold clamping process, injection unit advancement process, injection process, metering process, injection unit retreat process, mold opening process, eject process, and intermediate process are carried out in sequence.

First, the injection device 3 advances while the movable mold 28 is in close contact with the fixed mold 27, and the nozzle 31a of the injection device 3 is brought into close contact with the contact portion 22b, after which the injection process begins.

In the injection device 3, a molten molding material made of resin or low melting point metal is melted and kneaded in advance in the heating cylinder 31, and a predetermined amount is measured and stored in the front end of the heating cylinder 31. At that time, the flow path 22a is opened by the opening and closing valve 29, and the molten molding material is injected from the nozzle 31a at the same time. That is, when the screw in the heating cylinder 31 is given a forward movement and the measured molten molding material is injected from the nozzle 31a, the molten molding material flows into the cavity 27d. Since the molten molding material in the flow path 22a is heated to a predetermined temperature by the heating device 22c and its fluidity is maintained, the molten molding material in the flow path 22a is also filled into the cavity 27d formed in the center of the fixed mold 27 together with the molten molding material from the nozzle 31a.

Figure 5 is a flowchart showing the processing procedure of the processor 111 according to the first embodiment. The acquisition unit 114 of the molding condition adjustment device 101 acquires the detection value data output from the load cell 33a as injection pressure waveform data (step S111). The injection pressure waveform data is time series data of the injection pressure.

Next, the control unit 115 calculates difference information indicating the difference between the acquired injection pressure waveform data and the predetermined injection pressure waveform data (step S112). The difference information corresponds to the amount of deviation between the opening timing of the opening/closing valve 29 and the injection timing. The difference information can also be said to be information indicating, for example, a delay in the opening operation of the opening/closing valve 29.

Then, the control unit 115 refers to table 112b using the calculated difference information as a key, and identifies the amount of change in the open waiting time associated with the difference information (step S113).

Then, the control unit 115 changes the open waiting time included in the molding conditions based on the identified change amount (step S114).

Figure 6 is an explanatory diagram showing the molding condition adjustment method according to the first embodiment. Figure 6 shows a part of the molding cycle process. Figure 6A shows the state before the open waiting time is adjusted, and Figure 6B shows the state after the open waiting time is adjusted.

The "Open wait time" block indicates the wait time from the start of the molding process cycle (predetermined reference time) or the start of mold closing to the start of the open drive of the opening/closing valve 29. When the open wait time has elapsed, the control device 4 outputs a drive signal to the opening/closing valve 29 to start the open drive. The "Gate open time" block is the time required for the opening/closing valve 29 to operate and become open after the drive signal is output to the opening/closing valve 29. The width of the "Valve open time" block indicates the time until the opening/closing valve 29 actually operates and becomes open. The "Injection" block indicates the process in which injection is performed by the injection device 3.

If there is no problem with the opening and closing valve 29, the opening and closing valve 29 will be in the open state at the "gate opening time", so the "gate opening time" and the "valve opening time" should match. However, dirt on the sliding part of the opening and closing valve 29 may increase the time required for the opening operation. In Figure 6A, the dashed block portion of the "valve opening time" indicates that the opening operation is delayed. Since the injection of the molten molding material begins after the open waiting time and gate opening time have elapsed, if the opening operation of the opening and closing valve 29 is delayed, injection will begin with the opening and closing valve 29 not fully open, which may result in molding defects.

The molding condition adjustment device 101 executes the process shown in FIG. 5 to detect a delay in the opening operation of the opening/closing valve 29 and automatically adjusts the open wait time. The difference information described above corresponds to a delay in the opening operation of the opening/closing valve 29. If the opening timing of the opening/closing valve 29 is delayed, an injection pressure waveform like that shown in FIG. 4 is obtained. In this case, the amount of change to reduce the open wait time is specified in step S113, and the molding condition adjustment device 101 performs adjustment to reduce the open wait time. In other words, the open wait time is shortened.

The open wait time is shortened according to the length of the valve open time, and ultimately, as shown in FIG. 6B, the open wait time is adjusted so that the time when the opening/closing valve 29 actually opens coincides with the time when injection of the molten molding material starts. In other words, the length of the open wait time is adjusted so that the sum of the open wait time and the actual valve open time is always constant.

According to the injection molding machine 1 of the first embodiment, in an injection molding machine 1 using a mold 21 having a so-called valve gate, molding conditions can be automatically adjusted so that the opening timing of the opening/closing valve 29 matches the injection timing of the molten molding material. Specifically, the open waiting time of the opening/closing valve 29 can be automatically adjusted.

In addition, according to this embodiment 1, since the open waiting time is adjusted, the opening timing of the opening/closing valve 29 can be matched with the injection timing without affecting the cycle time of the molding process. In other words, the opening timing of the opening/closing valve 29 can be matched with the injection timing while keeping the sum of the open waiting time and the gate opening time shown in FIG. 6 constant. Also, there is no need to provide a margin in anticipation of a delay in the opening operation of the opening/closing valve 29, and the injection molding machine 1 can be operated in the shortest cycle.

(Embodiment 2)
The injection molding machine according to the second embodiment differs from that according to the first embodiment in the parameters of the molding conditions to be adjusted. Since the other configurations of the injection molding machine are similar to those of the injection molding machine according to the first embodiment, the same reference numerals are used for the similar parts, and detailed description thereof will be omitted.

Figure 7 is a conceptual diagram showing the configuration of table 112b according to embodiment 2. Table 112b according to embodiment 2 is information that associates difference information of the injection pressure waveform with the amount of change in the injection waiting time. The injection waiting time is the time from the end of the opening process of the opening/closing valve 29 to the start of injection of the molten molding material (see Figure 9). The open waiting time and gate opening time in embodiment 2 are fixed, predetermined times.

Figure 8 is a flowchart showing the processing procedure of the processor 111 according to the second embodiment. The acquisition unit 114 of the molding condition adjustment device 101 acquires injection pressure waveform data (step S211). Next, the control unit 115 calculates difference information indicating the difference between the acquired injection pressure waveform data and predetermined injection pressure waveform data (step S212).

Then, the control unit 115 refers to the table 112b using the calculated difference information as a key, and identifies the amount of change in the injection waiting time associated with the difference information (step S213). The control unit 115 then changes the injection waiting time included in the molding conditions based on the identified amount of change (step S214).

Figure 9 is an explanatory diagram showing the molding condition adjustment method according to embodiment 2. Figure 9 shows a part of the molding cycle process. Figure 9A shows the state before the injection wait time is adjusted, and Figure 9B shows the state after the injection wait time is adjusted. The "open wait time" and "gate open time" are the same as in embodiment 1. However, in embodiment 2, the "open wait time" is not subject to adjustment and is constant.

In Figure 9A, the dashed block portion of "valve open time" indicates that the opening operation is delayed. Since the injection of the molten molding material begins after the open waiting time and gate open time have elapsed, if the opening operation of the opening/closing valve 29 is delayed, injection will begin with the opening/closing valve 29 not fully open, which may result in molding defects.

The molding condition adjustment device 101 executes the process shown in FIG. 8 to detect a delay in the opening operation of the opening/closing valve 29 and automatically adjusts the injection wait time. If the opening timing of the opening/closing valve 29 is delayed, the amount of change to increase the injection wait time is identified in step S213, and the molding condition adjustment device 101 makes an adjustment to increase the injection wait time.

The injection wait time becomes longer depending on the length of the valve open time, and ultimately, as shown in Figure 9B, the open wait time is adjusted so that the time when the opening/closing valve 29 actually opens coincides with the time when injection of the molten molding material begins.
In addition, in FIG. 9A, if the time when the on-off valve 29 actually opens falls within the injection wait time and there is no effect on the quality of the molded product, it is not necessarily necessary to increase the injection wait time.

The injection molding machine 1 according to the second embodiment configured in this manner can automatically adjust the molding conditions, specifically the waiting time for the opening of the opening/closing valve 29, so that the opening timing of the opening/closing valve 29 coincides with the injection timing of the molten molding material, just like in the first embodiment.

In the second embodiment, an example in which only the ejection standby time is adjusted has been described, but the first and second embodiments may be combined to adjust either or both of the open standby time of the opening/closing valve 29 and the ejection standby time according to the difference information.

(Embodiment 3)
The injection molding machine according to the third embodiment differs from the first embodiment in that the molding conditions are adjusted by reinforcement learning. Since the other configurations of the injection molding machine are the same as those of the injection molding machine according to the first embodiment, the same reference numerals are used for the same parts, and detailed description thereof will be omitted.

Figure 10 is a block diagram showing an example of the configuration of a molding condition adjustment device 101 according to embodiment 3. The processor 111 of the molding condition adjustment device 101 according to embodiment 3 includes, as functional units, an acquisition unit 114, a control unit 115, and a reinforcement learning device 116. Each functional unit of the molding condition adjustment device 101 may be realized by software, or part or all of the functional units may be realized by hardware.

Figure 11 is a functional block diagram of the injection molding machine 1 according to the third embodiment. As shown in Figure 11, the control unit 115 has an observation unit 115a and a remuneration calculation unit 115b. The injection pressure waveform data acquired by the acquisition unit 114 is input to the observation unit 115a.

The observation unit 115a outputs the injection pressure waveform data acquired by the acquisition unit 114 and the molding conditions set in the control device 4 to the agent 116a of the reinforcement learning device 116 as observation data in reinforcement learning.

The reward calculation unit 115b calculates reward data from the difference between the acquired injection pressure waveform data and the predetermined injection pressure waveform data, and outputs the calculated reward data to the agent 116a of the reinforcement learning device 116. The reward data is similar to the difference information, and the smaller the deviation between the opening timing of the opening/closing valve 29 and the injection timing, the larger the reward value, and the larger the deviation, the smaller the reward value or the negative value.

As shown in FIG. 11, the reinforcement learner 116 includes an agent 116a and an adjustment unit 116b. The agent 116a is, for example, a reinforcement learning model having a deep neural network such as DQN, A3C, or D4PG, or a model-based reinforcement learning model such as PlaNet or SLAC.

In the case of a reinforcement learning model with a deep neural network, the agent 116a is equipped with a DQN (Deep Q-Network) and determines an action a according to the state s of the injection molding machine 1 indicated by the observation data, based on the state s.

The state s in this embodiment 3 includes at least information related to the acquired injection pressure waveform data. The state s may include information related to the injection pressure waveform data and molding conditions. For example, the state s may be image data represented by the acquired injection pressure waveform data, or may be difference information indicating the difference between the acquired injection pressure waveform data and the predetermined injection pressure waveform data. As described in the first embodiment, the difference information may be a numerical value indicating the difference between a coefficient that determines an approximation curve of the waveform indicated by the acquired injection pressure waveform data and a coefficient that determines an approximation curve of the waveform indicated by the predetermined injection pressure waveform data. The difference information may also be a numerical value indicating the area enclosed by the injection pressure waveform and the predetermined injection pressure waveform. The action a in this embodiment 3 is assumed to be a change amount parameter indicating the change amount of the open waiting time of the opening and closing valve 29.

DQN is a neural network model that outputs the value of each of multiple actions a when state s indicated by observed data is input. Multiple actions a correspond to multiple amounts of change in open waiting time. Actions a with high value represent appropriate amounts of change in open waiting time. Action a causes the injection molding machine 1 to transition to another state. After the state transition, agent 116a receives the reward calculated by reward calculation unit 115b, and agent 116a is trained to maximize the profit, i.e., the accumulated reward.

More specifically, DQN has an input layer, an intermediate layer, and an output layer. The input layer has a number of nodes to which the state s, i.e., observed data, is input. The output layer has a number of nodes that correspond to a number of actions a, respectively, and output the value Q(s, a) of the action a in the input state s.

Based on the state s, action a, and the reward r obtained by the action, the value Q expressed by the following formula (1) can be used as training data, and various weighting coefficients that characterize the DQN can be adjusted to enable reinforcement learning of the DQN of agent 116a.
Q(s, a)←Q(s,a)+α(r+γmaxQ(snext,anext)−Q(s,a))...(1)
however,
s: state a: action α: learning coefficient r: reward γ: discount rate maxQ(snext, next): maximum value of Q value for the next possible action

The adjustment unit 116b changes the open adjustment time set in the control device 4 based on the change amount (action a) of the open waiting time explored by the agent 116a during reinforcement learning, and outputs the changed change amount (action a) to the ejection device 3, the opening/closing valve 29, etc.

The reinforcement learning method according to the first embodiment will be described in detail below.
[Reinforcement learning processing]
It is assumed that an initial value of the open waiting time and other molding conditions are set in the injection molding machine 1, and a molding process cycle is repeated. The control unit 115 acquires the injection pressure waveform data output from the load cell 33a, and outputs observation data including the acquired injection waveform data and the molding conditions set in the control device 4 to the agent 116a of the reinforcement learning device 116.

The agent 116a of the reinforcement learning device 116 acquires the observation data output from the observation unit 115a. The agent 116a calculates a change amount parameter (action a) for adjusting the open waiting time of the injection molding machine 1 based on the observation data, and outputs the calculated change amount parameter (action a) to the adjustment unit 116b. The agent 116a selects the optimal action a during operation (inference), and during learning, it is preferable to determine an exploratory action a in order to reinforce learning the agent 116a. In addition, the agent 116a may select an action a with a small value of the objective function using an objective function in which the value is smaller the higher the action value or the more unexplored the action a is, and the value is larger the greater the change amount from the current molding conditions.

The adjustment unit 116b of the reinforcement learning device 116 changes the open waiting time based on the change amount parameter output from the agent 116a, and outputs the changed open waiting time to the ejection device 3, etc.

The injection molding machine 1 performs molding according to the adjusted open wait time and molding conditions. The load cell 33a measures the injection pressure in the injection molding machine 1 when the injection molding machine 1 performs molding, and the injection pressure waveform data obtained by the measurement is input to the observation unit 115a of the control unit 115 via the acquisition unit 114.

The observation unit 115a of the control unit 115 acquires the injection pressure waveform data, and outputs the acquired injection pressure waveform data and observation data including the molding conditions read from the control device 4 to the agent 116a of the reinforcement learning device 116. In addition, the reward calculation unit 115b calculates reward data based on the injection pressure waveform data, and outputs the calculated reward data to the reinforcement learning device 116.

The agent 116a updates the learning model based on the observation data output from the observation unit 115a and the reward data output from the reward calculation unit 115b. If the agent 116a is a DQN, the value expressed by the above formula (1) is used as training data to train the DQN.

The injection molding machine 1 according to the third embodiment can learn and automatically adjust the amount of change in the open waiting time of the on-off valve 29 by reinforcement learning so that the opening timing of the on-off valve 29 matches the injection timing of the molten molding material.

In the third embodiment, the agent 116a outputs a change amount parameter for the open waiting time of the opening/closing valve 29, but the agent 116a may be configured to output a parameter for the open waiting time of the opening/closing valve 29.

In addition, in this embodiment 3, a method for adjusting the open waiting time by reinforcement learning has been described, but similar to embodiment 2, the ejection waiting time may be configured to be adjusted by reinforcement learning.

Furthermore, although an example has been described in which the open waiting time or the injection waiting time of the opening/closing valve 29 is adjusted by reinforcement learning, the molding condition adjustment device 101 may be equipped with a learning device that has been trained by supervised learning.

The learning device is a model that outputs information for matching the opening timing of the opening/closing valve 29 with the injection timing of the molten molding material when molding conditions and injection pressure waveform data or difference information are input. Specifically, the learning device outputs the open wait time or injection wait time, or the amount of change in the open wait time or injection wait time, for matching the opening timing of the opening/closing valve 29 with the injection timing of the molten molding material.

The learning device may be a model having a convolution neural network (CNN) such as ResNet or DenseNet that recognizes the injection pressure waveform, or a model having a recurrent neural network (RNN), LSTM, Vision Transformer, etc. that recognizes the time-series transformation of the injection pressure.

The learning device may also be configured using a learning model with a configuration such as a neural network other than the above-mentioned CNN, RNN, etc., a Vision Transformer, an SVM (Support Vector Machine), a Bayesian network, or a decision tree such as XGBoost.

Furthermore, the molding condition adjustment device may include a first learning device and a second learning device. The first learning device adjusts the open waiting time or the injection waiting time when the value obtained by subtracting the integral value of the waveform related to the predetermined injection pressure waveform from the integral value of the waveform related to the acquired injection pressure waveform data is positive. The second learning device adjusts the open waiting time or the injection waiting time when the value obtained by subtracting the integral value of the waveform related to the predetermined injection pressure waveform from the integral value of the waveform related to the acquired injection pressure waveform data is negative. When the value obtained by the subtraction is positive, the molding condition adjustment device inputs the injection pressure waveform data to the first learning device. When the value obtained by the subtraction is negative, the molding condition adjustment device inputs the injection pressure waveform data to the second learning device. The molding condition adjustment device identifies and adjusts the amount of change in the open waiting time or the injection waiting time based on the output information of the first learning device and the second learning device.

REFERENCE SIGNS LIST 1 Injection molding machine 2 Clamping device 3 Injection device 4 Control device 5 Recording medium 21 Mold 22 Fixed platen 24 Movable platen 26 Clamping mechanism 27 Fixed mold 27d Cavity 28 Movable mold 29 Opening and closing valve 29a Needle 29b Air cylinder 31 Heating cylinder 31a Nozzle 32 Screw 33 Drive mechanism 33a Load cell 101 Molding condition adjustment device 111 Processor 112 Memory unit 112a Computer program 112b Table 113 Operation unit 114 Acquisition unit 115 Control unit 115a Observation unit 115b Reward calculation unit 116 Reinforcement learning device 116a Agent 116b Adjustment unit

Claims (16)

A molding condition adjustment method for adjusting molding conditions of an injection molding machine using a mold equipped with an opening/closing valve for opening and closing a flow path of a molten molding material leading to a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
by referring to a table in which the difference information is associated with a change amount of an open standby time from a predetermined reference point or a start point of mold closing in a molding process cycle to a start point of opening the on-off valve, a change amount of the open standby time corresponding to the calculated difference information is specified;
the open wait time is changed using the specified change amount, thereby adjusting the molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material coincide with each other. A molding condition adjustment method for adjusting molding conditions of an injection molding machine using a mold equipped with an opening/closing valve for opening and closing a flow path of a molten molding material leading to a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
By referring to a table in which the difference information is associated with a change amount of an injection waiting time from the end point of the opening process of the opening/closing valve, which is set a predetermined time after a predetermined reference point in a molding process cycle or a start point of mold closing, to the time when injection of a molten molding material starts, the change amount of the injection waiting time corresponding to the calculated difference information is specified;
By changing the injection waiting time using the specified change amount, the molding conditions are adjusted so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other.
How to adjust molding conditions. The difference information is
The area surrounded by the waveform represented by the acquired injection pressure waveform data and the waveform represented by the predetermined injection pressure waveform data.
The molding condition adjusting method according to claim 1 or 2 . The difference information is
is the difference between a coefficient that determines an approximation curve of the waveform indicated by the acquired injection pressure waveform data and a coefficient that determines an approximation curve of the waveform indicated by the predetermined injection pressure waveform data.
The molding condition adjusting method according to claim 1 or 2 . A molding condition adjustment method for adjusting molding conditions of an injection molding machine using a mold equipped with an opening/closing valve for opening and closing a flow path of a molten molding material leading to a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
When injection pressure waveform data is input, a learning device is used to output a change amount of the molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material, and the molding conditions are adjusted so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
How to adjust molding conditions. A molding condition adjustment method for adjusting molding conditions of an injection molding machine using a mold equipped with an opening/closing valve for opening and closing a flow path of a molten molding material leading to a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
When injection pressure waveform data is input, a reinforcement learning machine including an agent that outputs a change amount of molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material is used to adjust the molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
How to adjust molding conditions. Calculating remuneration data based on a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
The agent is made to undergo reinforcement learning based on the acquired injection pressure waveform data and the calculated reward data.
The molding condition adjusting method according to claim 6 . A computer program for causing a computer to execute a process for adjusting molding conditions of an injection molding machine using a mold having an opening/closing valve for opening and closing a flow path of a molten molding material that communicates with a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
by referring to a table in which the difference information is associated with a change amount of an open standby time from a predetermined reference point or a start point of mold closing in a molding process cycle to a start point of opening the on-off valve, a change amount of the open standby time corresponding to the calculated difference information is specified;
and changing the open wait time using the specified change amount, thereby adjusting molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material coincide with each other. A computer program for causing a computer to execute a process for adjusting molding conditions of an injection molding machine using a mold having an opening/closing valve for opening and closing a flow path of a molten molding material that communicates with a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
By referring to a table in which the difference information is associated with a change amount of an injection waiting time from the end point of the opening process of the opening/closing valve, which is set a predetermined time after a predetermined reference point in a molding process cycle or a start point of mold closing, to the time when injection of a molten molding material starts, the change amount of the injection waiting time corresponding to the calculated difference information is specified;
By changing the injection waiting time using the specified change amount, the molding conditions are adjusted so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other.
A computer program for causing the computer to execute a process. A computer program for causing a computer to execute a process for adjusting molding conditions of an injection molding machine using a mold having an opening/closing valve for opening and closing a flow path of a molten molding material that communicates with a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
When injection pressure waveform data is input, a learning device is used to output a change amount of the molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material, and the molding conditions are adjusted so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
A computer program for causing the computer to execute a process. A computer program for causing a computer to execute a process for adjusting molding conditions of an injection molding machine using a mold having an opening/closing valve for opening and closing a flow path of a molten molding material that communicates with a cavity, comprising:
Obtain injection pressure waveform data that shows the time change of injection pressure,
When injection pressure waveform data is input, a reinforcement learning machine including an agent that outputs a change amount of molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material is used to adjust the molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
A computer program for causing the computer to execute a process. A molding condition adjusting device that adjusts molding conditions of an injection molding machine using a mold having an opening and closing valve that opens and closes a flow path of a molten molding material that leads to a cavity, comprising:
An acquisition unit that acquires injection pressure waveform data indicating a time change of the injection pressure;
and a control unit that adjusts molding conditions so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other based on the acquired injection pressure waveform data .
The control unit is
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
by referring to a table in which the difference information is associated with a change amount of an open standby time from a predetermined reference point or a start point of mold closing in a molding process cycle to a start point of opening the on-off valve, a change amount of the open standby time corresponding to the calculated difference information is specified;
By changing the open waiting time using the specified change amount, the molding conditions are adjusted so that the opening timing of the on-off valve and the injection timing of the molten molding material coincide with each other.
Molding condition adjustment device. A molding condition adjusting device that adjusts molding conditions of an injection molding machine using a mold having an opening and closing valve that opens and closes a flow path of a molten molding material that leads to a cavity, comprising:
An acquisition unit that acquires injection pressure waveform data indicating a time change of the injection pressure;
a control unit that adjusts molding conditions so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other based on the acquired injection pressure waveform data;
Equipped with
The control unit is
Calculating difference information indicating a difference between the acquired injection pressure waveform data and a predetermined injection pressure waveform data;
By referring to a table in which the difference information is associated with a change amount of an injection waiting time from the end point of the opening process of the opening/closing valve, which is set a predetermined time after a predetermined reference point in a molding process cycle or a start point of mold closing, to the time when injection of a molten molding material starts, the change amount of the injection waiting time corresponding to the calculated difference information is specified;
By changing the injection waiting time using the specified change amount, the molding conditions are adjusted so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other.
Molding condition adjustment device. A molding condition adjusting device that adjusts molding conditions of an injection molding machine using a mold having an opening and closing valve that opens and closes a flow path of a molten molding material that leads to a cavity, comprising:
An acquisition unit that acquires injection pressure waveform data indicating a time change of the injection pressure;
a control unit that adjusts molding conditions so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other based on the acquired injection pressure waveform data;
Equipped with
The control unit is
When injection pressure waveform data is input, a learning device is used to output a change amount of the molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material, and the molding conditions are adjusted so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
Molding condition adjustment device. A molding condition adjusting device that adjusts molding conditions of an injection molding machine using a mold having an opening and closing valve that opens and closes a flow path of a molten molding material that leads to a cavity, comprising:
An acquisition unit that acquires injection pressure waveform data indicating a time change of the injection pressure;
a control unit that adjusts molding conditions so that the opening timing of the opening/closing valve and the injection timing of the molten molding material coincide with each other based on the acquired injection pressure waveform data;
Equipped with
The control unit is
When injection pressure waveform data is input, a reinforcement learning machine including an agent that outputs a change amount of molding conditions related to the opening drive of the on-off valve or the injection of the molten molding material in order to match the opening timing of the on-off valve with the injection timing of the molten molding material is used to adjust the molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material match.
Molding condition adjustment device. An injection molding machine using a mold having an opening and closing valve for opening and closing a flow path of a molten molding material leading to a cavity,
A molding condition adjusting device according to any one of claims 12 to 15 ,
an injection molding machine configured so that the molding condition adjustment device adjusts molding conditions so that the opening timing of the on-off valve and the injection timing of the molten molding material coincide with each other.

Images