JP7645110B2 - COMPUTER PROGRAM, INFORMATION PROCESSING METHOD, INFORMATION PROCESSING APPARATUS, AND INJECTION MOLDING SYSTEM - Google Patents

Description

The present invention relates to a computer program, an information processing method, an information processing device, and an injection molding system.

Patent document 1 discloses an injection molding scheduling system that automatically plans the molding sequence within a lot from basic data related to ordered products requested by customers, based on a pre-registered manufacturing process file.

Patent Document 2 discloses a schedule adjustment method for adjusting the injection process schedules of multiple injection molding machines connected to a network so that the maximum allowable power is not exceeded. In the schedule adjustment method, one injection molding machine connected to a network broadcasts its own injection process schedule to the other injection molding machines. The other injection molding machines receive the broadcasted injection process schedule and, based on the received injection process schedule, determine their own injection process schedule so that it falls within the maximum allowable power.

Patent document 3 discloses an online AI management system that stores molding condition programs for when various molding defects occur in a workstation connected to multiple injection molding machines, and supports the setting of molding conditions for each injection molding machine.

Patent document 4 discloses a state determination device that uses a machine learning device to determine abnormalities in an injection molding machine based on the operating state of injection molding.

Patent document 5 discloses a power management system that measures the power consumption of multiple injection molding machines, determines whether or not power reduction measures are necessary based on the maximum allowable power, and instructs the injection molding machines to implement power reduction measures as necessary.

Japanese Patent Application Publication No. 3-262622 JP 2011-121283 A Japanese Patent Application Publication No. 2-143825 JP 2018-167424 A JP 2010-240915 A

However, no technology has been disclosed that adjusts the settings of the acceleration, acceleration time, etc. of the motor of an injection molding machine while taking into account production volume and quality or power consumption.

The object of the present invention is to provide a computer program, information processing method, information processing device, and injection molding system that can maintain the required production volume and quality, or reduce demand power or power usage, and optimize production efficiency or power usage.

The computer program according to this embodiment is a computer program for causing a computer to execute a process of adjusting the setting values of each of a plurality of injection molding machines having a motor, and causes the computer to execute a process of acquiring status information obtained by observing at least one of the setting values of each of the plurality of injection molding machines, the quantity and quality of molded products produced by the plurality of injection molding machines, and the status related to the power consumed by the plurality of injection molding machines, determining an adjustment amount for the setting values including at least one of the acceleration, acceleration time, maximum speed, and constant speed operation time of the motors of the plurality of injection molding machines based on the acquired status information, and transmitting the determined adjustment amount for the setting values to the injection molding machines.

The information processing method according to this aspect is an information processing method for adjusting the setting values of each of a plurality of injection molding machines having a motor, and obtains status information obtained by observing at least one of the setting values of each of the plurality of injection molding machines, the quantity and quality of molded products manufactured by the plurality of injection molding machines, and the status related to the power consumed by the plurality of injection molding machines, and determines the amount of adjustment of the setting values including at least one of the acceleration, acceleration time, maximum speed, and constant speed operation time of the motors of the plurality of injection molding machines based on the obtained status information, and transmits the determined amount of adjustment of the setting values to the injection molding machines.

The information processing device according to this aspect is an information processing device that adjusts the setting values of each of a plurality of injection molding machines having a motor, and includes a status observation unit that observes at least one of the setting values of each of the plurality of injection molding machines, the quantity and quality of molded products manufactured by the plurality of injection molding machines, and a status related to the power consumed by the plurality of injection molding machines, a determination unit that determines an adjustment amount of the setting value including at least one of the acceleration, acceleration time, maximum speed, and constant speed operation time of the motor of the plurality of injection molding machines based on the status information observed by the status observation unit, and a transmission unit that transmits the adjustment amount of the setting value determined by the determination unit to the injection molding machine.

The injection molding system according to this aspect includes multiple injection molding machines each having a motor and the information processing device, and the information processing device transmits adjustment amounts for the setting values of the multiple injection molding machines to the injection molding machines, and the injection molding machines receive the adjustment amounts transmitted from the information processing device and adjust the setting values based on the received adjustment amounts.

As a result of the above, it is possible to maintain the required production quantity and quality, or reduce the demand power or power usage, thereby optimizing production efficiency or power usage.

1 is a block diagram showing an example of the configuration of an injection molding system according to an embodiment of the present invention. FIG. 2 is a block diagram showing a detailed configuration of an injection molding machine that constitutes the injection molding system. FIG. 2 is a block diagram showing a detailed configuration of a measurement unit constituting the injection molding system. FIG. 2 is an explanatory diagram conceptually illustrating the function of the injection molding system. FIG. 2 is a block diagram conceptually showing the configuration of a learning model. FIG. 2 is a conceptual diagram showing the power consumption of each process in a molding cycle in chronological order. 6 is a graph showing the operation and power consumption of an electric motor before adjustment of a set value. 11 is a graph showing the operation and power consumption of the electric motor after the setting value is adjusted. 13 is a flowchart showing a processing procedure of the information processing device. 13 is a flowchart showing a processing procedure of the information processing device. 13 is a flowchart showing a procedure for a remuneration calculation process. FIG. 13 is a conceptual diagram showing an example of an improvement in the operation end timing of two injection molding machines manufacturing different molded products. FIG. 13 is a schematic diagram showing an example of a monitor screen showing an adjustment status of a setting value.

Specific examples of a computer program, an information processing method, an information processing device, and an injection molding system according to embodiments of the present invention will be described below with reference to the drawings. At least some of the embodiments and variations described below may be combined in any manner. Note that the present invention is not limited to these examples, but is set forth in the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

1 is a block diagram showing an example of the configuration of an injection molding system according to this embodiment, FIG. 2 is a block diagram showing the detailed configuration of an injection molding machine 2 constituting the injection molding system, and FIG. 3 is a block diagram showing the detailed configuration of a measurement unit 4 constituting the injection molding system. The molding device system according to this embodiment includes an information processing device 1, a plurality of injection molding machines 2 installed in a factory, a factory power meter 3, and a measurement unit 4. The plurality of injection molding machines 2, the factory power meter 3, and the measurement unit 4 are connected to the information processing device 1 via a wired or wireless communication line, and the information processing device 1 transmits and receives various information between each injection molding machine 2, the factory power meter 3, and the measurement unit 4 to optimize molding conditions. The information processing device 1 may be configured to transmit and receive information with the measurement unit 4 via the control device 22 of the injection molding machine 2. In particular, the injection molding system according to this embodiment can reduce demand power and power usage while maintaining the required production quantity and quality by adjusting the settings such as acceleration, acceleration time, maximum speed, and constant speed operation time of the motor of the injection molding machine 2, and the heater temperature, thereby enabling optimization of production efficiency and power usage.

The information processing device 1 is a computer, and includes a processing unit 11, a storage unit 12, and a communication unit 13 as hardware configurations. The processing unit 11 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 processing unit 11 executes a computer program P stored in the storage unit 12 described later, thereby implementing the information processing method according to this embodiment, and functions as a state observation unit 11a, a determination unit 11b, a transmission unit 11c, a reward calculation unit 11d, and a reinforcement learning unit 11e (see FIG. 4). Details of each functional unit will be described later. Note that each functional unit of the information processing device 1 may be realized by software, or may be realized in part or in whole by hardware.
The information processing device 1 may be a server device connected to a network (not shown). The information processing device 1 may be configured to perform distributed processing using a plurality of computers, may be realized by a plurality of virtual machines provided in one server, or may be realized by using a cloud server.

The storage unit 12 is a non-volatile memory such as a hard disk, an EEPROM (Electrically Erasable Programmable ROM), a flash memory, etc. The storage unit 12 stores a computer program P for causing a computer to execute the information processing method according to the present embodiment. The storage unit 12 also stores a learning model 10 that performs reinforcement learning for adjusting the setting values of the injection molding machine 2. The learning model 10 will be described in detail later.
The computer program P according to the present embodiment may be recorded in a computer-readable manner on the recording medium 14. The storage unit 12 stores the computer program P read from the recording medium 14 by a reading device (not shown). The recording medium 14 is a semiconductor memory such as a flash memory. The recording medium 14 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 14 may also be a magnetic disk such as a flexible disk or a hard disk, a magnetic optical disk, or the like. Furthermore, the computer program P according to the present embodiment may be downloaded from an external server (not shown) connected to a communication network (not shown) and stored in the storage unit 12.

The communication unit 13 is a communication circuit that transmits and receives information according to a predetermined communication protocol such as Ethernet (registered trademark). The communication unit 13 is connected to multiple injection molding machines 2, factory power measuring devices 3, and measurement units 4 via a wired or wireless communication network, and the processing unit 11 can transmit and receive various information between the injection molding machines 2, factory power measuring devices 3, and measurement units 4 via the communication unit 13. The communication unit 13 may be configured to transmit and receive information to and from the measurement units 4 via the control device 22.

The factory power meter 3 measures the power consumed by the multiple injection molding machines 2, the information processing device 1, and various other electrical devices in the factory, and transmits information indicating the total power consumption consumed in the factory to the information processing device 1. The total power consumption is the information used to calculate the demand power (demand value) in the factory. The demand power is a value measured as the average power consumption for a specified demand time (30 minutes). The power fee consists of a basic fee and a power fee determined by the amount of power actually used, and the demand power is used as the basis for calculating this basic fee. Specifically, the largest demand power in a month is the maximum demand value for that month, and the basic fee is determined, for example, by the maximum demand value for the most recent 12 months. In order to reduce the demand value that affects the basic fee, it is not sufficient to only monitor the power consumption of the injection molding machines 2, so the factory power meter 3 measures the total power consumption in the entire factory.

The injection molding machine 2 includes an injection device body (not shown) and a mold clamping device disposed in front of the injection device body. The injection device body includes a heating cylinder, a heater 20 for heating the heating cylinder, a hopper for feeding resin material into the heating cylinder, a screw that is provided in the heating cylinder so as to be drivable in the rotational direction and the axial direction, an injection electric motor 21a for driving the screw in the axial direction, and a metering and plasticization electric motor 21b for driving the screw in the rotational direction. The mold clamping device includes a toggle mechanism that opens and closes the mold and clamps the mold so that it does not open when the molten resin injected from the injection device body is filled into the mold, a mold opening and closing electric motor 21c that drives the toggle mechanism, and an ejection electric motor 21d for ejecting the molded product from the mold. Hereinafter, the electric motor 21a for injection, the electric motor 21b for metering and plasticization, the electric motor 21c for opening and closing the mold, and the electric motor 21d for ejection will be collectively referred to as the electric motor 21.

The injection molding machine 2 also includes a control device 22 that controls the operation of the injection device body. The control device 22 accepts operations by an operator and controls the operation of the injection molding machine 2. The operator can input set values that determine the molding conditions of the injection molding machine 2 by operating an operation panel of the control device 22. The set values that are molding conditions include, for example, in-mold resin temperature, cylinder temperature, mold clamping force, injection speed, injection acceleration, injection peak pressure, injection stroke, 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, metering completion position, screw retreat speed, cycle time, intermediate time (waiting time between molding cycles), mold closing time, injection time, holding pressure time, metering time, mold opening time, etc.
In particular, the set values to be adjusted in this embodiment include, for example, the acceleration, acceleration time, maximum speed, and constant speed operation time of the electric motor 21a for injection, the electric motor 21b for metering and plasticization, the electric motor 21c for mold opening and closing, and the electric motor 21d for ejection. Note that the acceleration and acceleration time include negative acceleration and acceleration time, that is, the deceleration and deceleration time of each electric motor 21. The set values to be adjusted in this embodiment also include the heater temperature. Furthermore, the intermediate time, which is the waiting time between molding cycles, the driving carrier frequency of the semiconductor that drives the electric motor 21, and the downtime of the electric motor 21 may also be settings to be adjusted.

Furthermore, the control device 22 receives an allowable adjustment range of the set value according to the information processing method of this embodiment. The allowable adjustment range is information indicating the upper and lower limits of the amount of adjustment of the set value that can be automatically adjusted by the information processing device 1. The operator can input the allowable adjustment range of the set value by operating the operation panel of the control device 22.
The control device 22 transmits to the information processing device 1 information indicating basic information of the molded product, the set value of the injection molding machine 2, the adjustment allowable range of the set value, the power consumption in the injection molding machine 2, the temperature of the heating cylinder, information indicating the quantity and quality of the molded product manufactured per unit time, and the identifier of the injection molding machine 2. The information processing device 1 receives information transmitted from the control device 22 as described later, calculates an adjustment amount of the set value of each injection molding machine 2 that reduces the peak power, the demand power, and the average power consumption while maintaining the required production quantity and quality based on the received information, and transmits the calculated adjustment amount to the control device 22 of each injection molding machine 2. The control device 22 receives the adjustment amount transmitted from the information processing device 1, and adjusts the set values of the intermediate time, the heater temperature, the acceleration of each electric motor 21, the acceleration time, the maximum speed and the constant speed operation time, the pause time, the semiconductor drive carrier frequency, etc. based on the received adjustment amount. The control device 22 executes reception of the adjustment amount and adjustment control in real time.

The measurement unit 4 is a measuring device that measures physical quantities related to the state of the injection unit body and the molded product. The physical quantities include temperature, position, speed, acceleration, current, voltage, pressure, time, image data, torque, force, distortion, power consumption, etc. In particular, the measurement unit 4 according to this embodiment includes a molding machine power measurement unit 4a, a temperature measurement unit 4b, a quantity measurement unit 4c, a quality measurement unit 4d, etc.

The molding machine power measurement unit 4a measures the power consumption of the injection molding machine 2 and outputs the information obtained by the measurement to the control device 22. The temperature measurement unit 4b measures the temperature of the heating cylinder and transmits the information obtained by the measurement to the information processing device 1. The quantity measurement unit 4c measures the quantity molded by the injection molding machine 2 per unit time and transmits the information obtained by the measurement to the information processing device 1. The quantity measurement unit 4c includes, for example, an imaging device, an optical sensor, etc., measures the quantity of molded products, and transmits the information obtained by the measurement to the information processing device 1. The quality measurement unit 4d specifies the quality of the molded products and transmits the information obtained by the measurement to the information processing device 1. The quality measurement unit 4d includes, for example, an imaging device, an optical measuring instrument such as a laser displacement sensor, a weight scale, a strength measuring instrument, etc.
The quality measuring unit 4d can identify quality abnormalities related to burrs, sink marks, and other deformations of the molded product, for example, based on a camera image obtained by capturing an image of the molded product and the amount of deformation of the molded product obtained by a laser displacement sensor. The quality measuring unit 4d can also identify the quality of the molded product based on optical measurement values such as the chromaticity and brightness of the molded product obtained by an optical measuring instrument, the weight of the molded product measured by a weight scale, the strength of the molded product measured by a strength measuring instrument, and the like.
The molding machine power measurement unit 4a, the temperature measurement unit 4b, the quantity measurement unit 4c, or the quality measurement unit 4d may output information obtained by measurement to the control device 22 and transmit the information to the information processing device 1 via the control device 22. The measurement unit 4 may be built into or mounted on the injection molding machine 2.

FIG. 4 is an explanatory diagram conceptually showing the functions of the injection molding system. Each functional unit and the learning model 10 of the information processing device 1 will be described. The state observation unit 11a acquires information transmitted from the information processing device 1 and the factory power meter 3. Specifically, the state observation unit 11a acquires a plurality of set values of the injection molding machine 2 transmitted from the information processing device 1, the adjustment allowable range of each set value, the power consumption in the injection molding machine 2, the temperature of the heating cylinder, information indicating the quantity and quality of molded products manufactured per unit time, and the identifier of the injection molding machine 2. In addition, the state observation unit 11a acquires information indicating the total power consumption of the factory transmitted from the factory power meter 3. Furthermore, the state observation unit 11a acquires target conditions. The target conditions include, for example, target quantity conditions indicating the target quantity of molded products to be manufactured within a predetermined period, molded product quality conditions required for the molded products, target demand power in the factory, target peak power, and target average power. The target conditions may be obtained from the injection molding machine 2, or may be obtained from another device that manages the operation of multiple injection molding machines 2, or may be input to the information processing device 1.
The status observation unit 11a provides various information acquired from the injection molding machine 2, the factory power measuring device 3, etc. as status information to the learning model 10 and the reward calculation unit 11d.

Figure 5 is a block diagram conceptually illustrating the configuration of the learning model 10. The learning model 10 is, for example, a DQN (Deep Q Network). The learning model 10 has an input layer 10a to which state information provided by the state observation unit 11a is input, an intermediate layer 10b that extracts features of the state information, and an output layer 10c that outputs value information related to the adjustment amount of the setting value of each of the multiple injection molding machines 2. The value information is, for example, the value when an action is taken to adjust a certain setting value by a specific adjustment amount in a certain state, that is, a state action value (Q value).

The input layer 10a has multiple nodes where target conditions, settings of multiple injection molding machines 2, power information, quantity of molded products, quality of molded products, heater temperature, etc. are input, and passes the input information to the intermediate layer 10b. The target conditions are, for example, the target quantity of molded products to be produced in a factory within a specified period of time, the quality of the molded products required for the molded products, the target demand power in the factory, the target peak power, the target average power, etc. The power information includes the total power consumption in the factory and the power consumption of each of the multiple injection molding machines 2. The quantity and quality of molded products and heater temperature input to the input layer 10a are the quantity and quality of molded products produced by each of the multiple injection molding machines 2, and the heater temperature of each of the multiple injection molding machines 2.

The intermediate layer 10b compresses the state information input to the input layer 10a and extracts the features of the current state of the injection molding system. For ease of drawing, the intermediate layer 10b is illustrated as a single block, but the intermediate layer 10b is made up of multiple layers, and each layer has multiple nodes.

The output layer 10c has multiple nodes corresponding to the adjustment amounts of the set values in the multiple injection molding machines 2, and each node outputs value information of the corresponding adjustment amount. There are multiple set values set in the multiple injection molding machines 2, and there are also multiple adjustment amounts for each set value. In the example shown in FIG. 5, the node corresponding to the adjustment amount +ΔA for the set value A of the first injection molding machine 2 outputs value information for adjusting the set value A by the adjustment amount +ΔA. Similarly, the node corresponding to the adjustment amount -ΔB for the set value B outputs value information for adjusting the set value B by the adjustment amount -ΔB. Similarly, value information is output in the same manner for the multiple set values of the second injection molding machine 2. The set values for which value information is output from the learning model 10 include the acceleration, acceleration time, maximum speed and constant speed operation time, and heater temperature of the injection electric motor 21a, the metering and plasticization electric motor 21b, the mold opening and closing electric motor 21c, and the ejection electric motor 21d of the multiple injection molding machines 2. Additionally, the set values for which value information is output from the learning model 10 may include intermediate time, the driving carrier frequency of the semiconductor that drives the electric motor 21, and the downtime of the electric motor 21.

The determination unit 11 shown in FIG. 4 determines the adjustment amount of the set value of each of the multiple injection molding machines 2 based on the value information output from the learning model 10. Specifically, among the adjustment amounts of the set values of each of the multiple injection molding machines 2, the determination unit 11b determines the set value and adjustment amount corresponding to the node that outputs the value information with the largest Q value as the set value and adjustment amount to be adjusted. However, the determination unit 11b determines the adjustment amount so that the adjustment amount of the set value is within the adjustment allowable range. Note that the determination unit 11b may be configured to determine one set value and adjustment amount with the largest Q value, or to determine a predetermined number of set values and adjustment amounts with the largest Q value, or to determine the set value and adjustment amount for the prioritized drop in order from the set value of the injection molding machine 2 with the largest production quantity and power consumption, or to determine one set value and adjustment amount for each injection molding machine 2, or to determine one set value and adjustment amount for each electric motor 21 of the multiple injection molding machines 2.
The transmitting unit 11c transmits the adjustment amount determined by the determining unit 11b, that is, the adjustment amount of the setting value of each of the multiple injection molding machines 2, to the control device 22 of each injection molding machine 2.

The reward calculation unit 11d calculates the reward based on the status information provided by the status observation unit 11a. If the quantity and quality of the molded products manufactured by the injection molding machine 2 satisfy the target quantity condition and the target quality condition, the reward is added, and if the target quantity condition and the target quality condition are not satisfied, the reward is subtracted. Furthermore, the reward calculation unit 11d adds the reward when the demand power is smaller than the target demand power, and subtracts the reward when it is larger than the target demand power. Furthermore, the reward calculation unit 11d adds the reward when the peak power or average power consumed by the multiple injection molding machines 2 is smaller than the target peak power and the target average power, and subtracts the reward when it is larger than the target peak power and the target average power.

Based on the reward output from the reward calculation unit 11d, the reinforcement learning unit 11e trains the learning model 10 by machine learning using the state action value Q calculated before the adjustment of the set value and the TD error obtained from the reward after the adjustment of the set value. The value update formula is expressed by the following formula (1).

Q(st, at)←Q(st,at)+α(R+γmaxQ(st+1,at+1)−Q(st,at))…(1)
however,
st: state at time t; at: action at time t (adjustment of set value);
Q(st, at): State action value α: Learning coefficient R: Immediate reward γ: Time discount rate maxQ(st+1, at+1): State action value when selecting the action with the maximum value for the next step

The reinforcement learning unit 11e optimizes the weighting coefficients of the learning model 10 by using backpropagation, gradient descent, or the like so that the TD error, i.e., the second term on the right-hand side of the above formula (1) (TD error), approaches zero, thereby training the learning model 10 through machine learning.

Although Q-learning has been described as a machine learning method, other value iteration methods such as SARS A may be used. Furthermore, policy gradient methods may be used as machine learning methods. Furthermore, although DQN has been described as the learning model 10, a table that associates state information with action values may be used.

Figure 6 is a conceptual diagram showing the power consumption of each process in a molding cycle over time. The horizontal axis shows time, and the vertical axis shows power consumption. The injection molding process mainly includes (1) mold closing process, (2) injection process, (3) mold opening process, (4) metering (and plasticization) process, and (5) ejection process, and processes (1) to (5) are performed consecutively to constitute an injection molding cycle. Figure 6 shows the change in power consumption of each process over time in two cycles.

(1) The mold closing process is a process in which the mold is closed by the mold clamping device, and the mold opening and closing electric motor 21c is driven.
(2) The injection process is a process in which the measured and plasticized resin is injected into a mold, and is driven by the injection electric motor 21a.
(3) The mold opening process is a process of opening the mold to remove the molded product that has been cooled inside the mold, and is driven by the electric motor 21c for opening and closing the mold.
(4) The metering process is a process performed in parallel with the mold opening process to measure and plasticize the resin to be injected in the next injection molding cycle, and is driven by the electric motor 21b for metering and plasticization.
(5) The ejection process is a process in which, after the mold is opened, the molded product is removed from the mold by an ejection pin, and is driven by the ejection electric motor 21d.

Figure 7 is a graph showing the power consumption of the electric motor 21 before the setting value is adjusted, and Figure 8 is a graph showing the power consumption of the electric motor 21 after the setting value is adjusted. The horizontal axis shows time, and the vertical axis shows power consumption. As shown in Figures 7 and 8, the power consumption of the electric motor 21 can be changed by the motor's acceleration, acceleration time, maximum speed, and constant speed operation time.

Figures 9 and 10 are flowcharts showing the processing procedure of the information processing device 1. During operation of the multiple injection molding machines 2, the information processing device 1 adjusts the setting values of each injection molding machine 2 in real time by executing the following processes. The information processing device 1 acquires target conditions (step S11). The information processing device 1 acquires information indicating the setting values and the adjustment tolerance range from the injection molding machines 2 (step S12). The information processing device 1 acquires information indicating the power consumption of each injection molding machine 2, information indicating the quantity and quality of molded products, and information indicating the heater temperature from the multiple injection molding machines 2 (step S13). The information processing device 1 acquires information indicating the total power consumption from the factory power meter 3 (step S14).

The processing unit 11 inputs the acquired target conditions, the setting value of each injection molding machine 2, power consumption, the quantity and quality of the molded products, the heater temperature, and the total power consumption of the factory into the learning model 10, thereby calculating the value of the adjustment amount of the setting value of each injection molding machine 2 (step S15). The processing unit 11 determines the setting value and adjustment amount to be adjusted based on the calculated value of each adjustment amount and the adjustment allowable range (step S16). The processing unit 11 determines the adjustment amount of the setting value that is within the adjustment allowable range and has a large value as the adjustment target. Next, the processing unit 11 transmits the determined adjustment amount to one or more injection molding machines 2 (step S17). The control device 22 of the injection molding machine 2 receives the adjustment amount transmitted from the information processing device 1, changes the setting value based on the received adjustment amount, and controls the molding process of the injection molding machine 2 using the changed setting value.

Next, after adjusting the set values, the processing unit 11 of the information processing device 1 acquires information indicating the power consumption of each injection molding machine 2, information indicating the quantity and quality of molded products, and information indicating the heater temperature from the multiple injection molding machines 2 (step S18). The information processing device 1 acquires information indicating the total power consumption from the factory power meter 3 (step S19). The processing unit 11 then calculates the remuneration based on the various acquired status information (step S20).

Figure 11 is a flowchart showing the procedure for calculating the reward. For the sake of explanation, the following will be explained in separate steps for each element to which the reward is added or subtracted, but it goes without saying that the reward may also be calculated in one step using a function or table.

The processing unit 11 adds or subtracts the reward according to the actually measured demand power of the factory and the peak power and average power based on the power consumption of each injection molding machine 2 relative to the target power (step S51). Specifically, the processing unit 11 adds the reward when the measured demand power is equal to or less than the target demand power, and subtracts the reward when the measured demand power exceeds the target demand power. Similarly, the processing unit 11 adds the reward when the peak power and average power are equal to or less than the target peak power and target average power, and subtracts the reward when the target peak power and target average power are exceeded.

The processing unit 11 also adds or subtracts the reward depending on the quantity of molded products relative to the target quantity condition (step S52). Specifically, if the quantity of molded products manufactured by the injection molding machine 2 satisfies the target quantity condition, the reward is added, and if the target quantity condition is not satisfied, the reward is subtracted.

Furthermore, the processing unit 11 adds or subtracts the reward depending on the quality of the molded product relative to the target quality conditions (step S53). Specifically, if the quality of the molded product manufactured by the injection molding machine 2 satisfies the target quality conditions, the reward is added, and if the quality of the molded product does not satisfy the target quality conditions, the reward is subtracted.

Furthermore, the processing unit 11 adds or subtracts the reward according to the measured heater temperature relative to the target heater temperature (step S54). Specifically, if the actually measured heater temperature of the injection molding machine 2 deviates from the target heater temperature, the reward is subtracted.

Returning to FIG. 9, after the processing unit 11 has finished calculating the reward, it performs reinforcement learning on the learning model 10 based on the calculated reward so as to reduce, for example, the TD error (step S21).

Then, the processing unit 11 determines whether or not a predetermined processing end condition is satisfied (step S22). If it is determined that the end condition is satisfied (step S22: YES), the processing unit 11 ends the processing. If it is determined that the processing end condition is not satisfied (step S22: YES), the processing unit 11 returns the processing to step S13.

By the above process, the setting values of each injection molding machine 2 are adjusted in real time during the molding process in multiple injection molding machines 2, optimizing the molding conditions. Specifically, it is possible to reduce demand power, peak power, and average power while maintaining the required production volume and quality, optimizing production efficiency and power usage.

Fig. 12 is a conceptual diagram showing an example of an improvement in the operation end timing of two injection molding machines 2 that manufacture different molded products. In the example shown in Fig. 12, two injection molding machines 2 are operating, and in a given operating time, the first injection molding machine 2 operates for three cycles and the second injection molding machine 2 operates for five cycles. The two injection molding machines 2 manufacture different types of molded products.
12 shows the molding cycle before the adjustment of the set value, and the lower diagram shows the molding cycle before the adjustment of the set value. When the acceleration, acceleration time, intermediate time, etc. of each electric motor 21 are adjusted so that the demand power, peak power, and average power are reduced while maintaining the required production quantity and quality, the cycle time of each injection molding machine 2 is naturally lengthened. Then, when it is required that the first and second injection molding machines 2 manufacture the required quantity of molded products at a predetermined operation end time, the set value of the electric motor 21 is adjusted so that the operation end times of the first and second injection molding machines 2 are aligned. In this way, the production of the required quantity of molded products by the first injection molding machine 2 and the production of the required quantity of molded products by the second injection molding machine 2 end at the same time, which contributes to reducing excess inventory.

The information processing device 1 may be configured to display the adjustment status of the setting values of the multiple electric motors 21 of the multiple injection molding machines 2 on a display device (not shown).

13 is a schematic diagram showing an example of a monitor screen showing the adjustment status of the set values. The monitor screen has an operation status display section 91 showing the operation status of the multiple injection molding machines 2, and an adjustment amount display section 92 showing the adjustment amount of the set value in each injection molding machine 2.
The operation status display unit 91 displays the operation status before the adjustment of the set values with a solid line, and the operation status after the adjustment of the set values with a dashed line. The operation status is displayed, for example, by the cycle times and intermediate times of the multiple injection molding machines 2 as the lengths of arrow images and the intervals between the arrow images.
The adjustment amount display unit 92 displays the set values and adjustment amounts of the acceleration, acceleration time, maximum speed, and constant speed operating time of each of the electric motors 21a for injection, 21b for metering and plasticization, 21c for mold opening and closing, and 21d for ejection in each injection molding machine 2.

The control device 22 of the injection molding machine 2 may also be configured to display the settings and adjustment amounts of the machine itself.

According to the injection molding system of this embodiment configured in this manner, by adjusting the settings for the acceleration, acceleration time, maximum speed or constant speed operation time, heater temperature, intermediate time, etc. of the electric motors 21 of the multiple injection molding machines 2, it is possible to reduce the demand power and power usage while maintaining the required production quantity and quality, thereby optimizing production efficiency and power usage.

Specifically, by adjusting the detailed settings such as the acceleration, acceleration time, maximum speed and constant speed operation time, and heater temperature of the injection motor 21a, the metering and plasticizing motor 21b, the mold opening and closing electric motor 21c, and the ejection electric motor 21d of the injection molding machine 2, production efficiency and power usage can be optimized.

In addition, the end points of the molding cycles of each injection molding machine 2 are adjusted to coincide at the end of the operating time, which can reduce excess inventory.

Furthermore, the information processing device 1 can perform reinforcement learning of how to adjust the setting values, and adjust the setting values of the injection molding machine 2 in real time. In particular, by using DQN, it is possible to appropriately adjust multiple setting values in multiple injection molding machines 2.

Furthermore, the information processing device 1 is configured to adjust the settings within the adjustment tolerance range of each setting value set by the operator, so that production efficiency and power consumption can be optimized through reinforcement learning while ensuring stable operation of the injection molding machine 2.

In the present embodiment, an example has been described in which the injection molding system automatically adjusts the set values, but the final adjustment of the set values may be performed by an operator. For example, the control device 22 of the injection molding machine 2 may be configured to display the adjustment amount transmitted from the information processing device 1 to the injection molding machine 2, accept the adjustment and decision of the set values from the operator, and change the set values according to the accepted adjustment amount. In addition, some set values may be manually adjusted by the operator. Furthermore, an example has been described in which the demand power and power usage are reduced while maintaining the required production quantity and quality, but the production efficiency or power usage may be optimized based on at least one of the production quantity and quality and the power consumption.

REFERENCE SIGNS LIST 1 Information processing device 2 Injection molding machine 3 Factory power measuring device 10 Learning model 10a Input layer 10b Intermediate layer 10c Output layer 11 Processing unit 11a State observation unit 11b Determination unit 11c Transmission unit 11d Reward calculation unit 11e Reinforcement learning unit 12 Memory unit 13 Communication unit 14 Recording medium 20 Heater 21 Electric motor 21a Injection electric motor 21b Metering and plasticization electric motor 21c Mold opening and closing electric motor 21d Ejection electric motor 22 Control device 4 Measurement unit 4a Molding machine power measurement unit 4b Temperature measurement unit 4c Quantity measurement unit 4d Quality measurement unit 9 Monitor screen 91 Operation status display unit 92 Adjustment amount display unit P Computer program

Claims (11)

A computer program for causing a computer to execute a process of adjusting setting values of each of a plurality of injection molding machines having an electric motor for injection, an electric motor for metering and plasticization, an electric motor for opening and closing a mold, and an electric motor for ejecting a molded product, the computer program comprising:
Obtaining status information obtained by observing the setting values of each of the multiple injection molding machines, the quantity and quality of molded products manufactured by the multiple injection molding machines, and the status related to the power consumed by the multiple injection molding machines;
determining adjustment amounts of set values including acceleration, acceleration time, maximum speed, and constant speed operation time of each electric motor of the plurality of injection molding machines based on the acquired status information;
and transmitting the determined adjustment amount of the set value to the injection molding machine. When the state information is input, the acquired state information is input to a learning model that outputs value information related to an adjustment amount of a setting value, thereby outputting value information;
The computer program product according to claim 1 , for causing the computer to execute a process of determining an adjustment amount of a setting value based on value information output from the learning model. Calculating a reward for the adjustment of the setting value based on the state information obtained after transmitting the determined adjustment amount of the setting value;
The computer program product according to claim 2 , for causing the computer to execute a process of performing reinforcement learning on the learning model based on the calculated reward. The computer program according to claim 3, which causes the computer to execute a process of adding a reward when the quantity of molded products satisfies a predetermined target quantity condition, and subtracting the reward when the quantity of molded products does not satisfy the target quantity condition. The computer program according to claim 3 or claim 4, which causes the computer to execute a process of adding a reward when the quality of the molded product satisfies a predetermined quality condition, and subtracting the reward when the quality of the molded product does not satisfy the quality condition. The computer program according to any one of claims 3 to 5, for causing the computer to execute a process of increasing a reward as a demand power of a facility having a plurality of the injection molding machines and a peak power or an average power consumed by the plurality of the injection molding machines are lower. the injection molding machine includes a heater;
7. The computer program according to claim 1, for causing the computer to execute a process of determining adjustment amounts of set values including an acceleration, an acceleration time, a maximum speed, a constant speed operation time, and a temperature of the heater of each electric motor of the injection molding machine. acquiring information indicating an adjustment allowable range of a setting value of the injection molding machine;
The computer program product according to claim 1 , which causes the computer to execute a process of determining an adjustment amount of a set value within the acquired adjustment allowable range. 1. An information processing method for adjusting setting values of a plurality of injection molding machines each having an electric motor for injection, an electric motor for metering and plasticization, an electric motor for opening and closing a mold, and an electric motor for ejecting a molded product, comprising:
Obtaining status information obtained by observing the setting values of each of the multiple injection molding machines, the quantity and quality of molded products produced by the multiple injection molding machines, and the status related to the power consumed by the multiple injection molding machines;
determining adjustment amounts of set values including acceleration, acceleration time, maximum speed, and constant speed operation time of each electric motor of the plurality of injection molding machines based on the acquired status information;
The determined adjustment amount of the set value is transmitted to the injection molding machine. An information processing device for adjusting setting values of a plurality of injection molding machines each having an electric motor for injection, an electric motor for metering and plasticization, an electric motor for opening and closing a mold, and an electric motor for ejecting a molded product, comprising:
a status observation unit that observes the setting values of each of the multiple injection molding machines, the quantity and quality of molded products manufactured by the multiple injection molding machines, and a status related to power consumed by the multiple injection molding machines;
a determination unit that determines adjustment amounts of set values including acceleration, acceleration time, maximum speed, and constant speed operation time of each electric motor of the plurality of injection molding machines based on state information observed by the state observation unit;
a transmission unit that transmits the adjustment amount of the setting value determined by the determination unit to the injection molding machine. a plurality of injection molding machines each having an electric motor for injection, an electric motor for metering and plasticization, an electric motor for opening and closing a mold, and an electric motor for ejecting a molded product;
The information processing device according to claim 10,
The information processing device transmits adjustment amounts of setting values of the multiple injection molding machines to the injection molding machines, and the injection molding machines receive the adjustment amounts transmitted from the information processing device and adjust the setting values based on the received adjustment amounts.

Images