Information processing device, information processing method, information processing program and manufacturing method for molded products

Abstract

To provide an information processing device, an information processing method, an information processing program, and a manufacturing method for molded products that accurately predict the quality of molded products.SOLUTION: An information processing device (2) includes a condition acquisition unit (201) that acquires physical property conditions that indicate the physical properties of a molded product manufactured by blending raw materials extracted from microorganisms and process values of the raw materials calculated by a simulation simulating an extruder, and a quality prediction unit (202) that predicts the quality using a quality prediction model (211) that uses the physical property conditions and the process values as explanatory variables and the quality of the molded product as a target variable.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, an information processing program, and a method for manufacturing a molded product. [Background technology]

[0002] A method is known in which glass fibers and various additives are blended with polyphenylene sulfide resin to form a fiber-reinforced polyphenylene sulfide resin composition. While the blending ratios and other properties of such fiber-reinforced polyphenylene sulfide resin compositions have traditionally been determined by trial and error, the invention described in Patent Document 1 uses a learning model to predict optimal blending ratios and other properties. The invention described in Patent Document 1 makes it possible to maximize the strength and minimize the linear expansion coefficient of molded products. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-51839 Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, environmental problems caused by discarded plastics have been attracting attention. In particular, marine pollution caused by discarded plastics is serious, and there are high hopes for the widespread use of biodegradable polymers that decompose in the natural environment.

[0005] An example of a biodegradable polymer is poly(3-hydroxybutyrate-co-3-hydroxyhexanoate), a copolymer of 3-hydroxybutyrate and 3-hydroxyhexanoate. Such biodegradable polymers are accumulated as energy storage substances within the cells of microbial species and can biodegrade not only in soil but also in seawater, making them attractive as materials that can solve the above problems.

[0006] Because the production process of biodegradable polymers is based on biochemical processes, differences in physical properties can occur between culture lots, even for the same type of polymer. Therefore, switching between culture lots can have a negative impact on the quality of molded products (e.g., pellets).

[0007] Therefore, it is possible to predict the quality of molded products using a learning model such as that described in Patent Document 1. One example of an explanatory variable for the learning model is "resin temperature," which is one of the production conditions that affect the quality of molded products. However, because resin temperature is typically measured only near the tip of the extruder, it cannot be said to fully reflect the effects of production conditions. Furthermore, resin temperature is measured at a single point on the extruder using a thermocouple, and there is a risk of unexpected deterioration in measurement accuracy due to changes in the resin filling rate or sensor degradation. Therefore, in order to accurately predict the quality of molded products, an explanatory variable that can replace resin temperature is required.

[0008] An object of one aspect of the present invention is to accurately predict the quality of a molded product. [Means for solving the problem]

[0009] In order to solve the above problems, an information processing device according to one embodiment of the present invention includes a condition acquisition unit that acquires physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that mimics an extruder used to manufacture the molded product, and a quality prediction unit that predicts the quality of the molded product from the physical property conditions and the process values ​​using a quality prediction model that uses the physical property conditions and the process values ​​acquired by the condition acquisition unit as explanatory variables and the quality of the molded product as a target variable.

[0010] In order to solve the above-mentioned problems, an information processing method according to one embodiment of the present invention is an information processing method executed by one or more information processing devices, and includes a condition acquisition step of acquiring physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that mimics an extruder used to manufacture the molded product, and a quality prediction step of predicting the quality of the molded product from the physical property conditions and the process values ​​using a quality prediction model that uses the physical property conditions and the process values ​​as explanatory variables and the quality of the molded product as a target variable.

[0011] In order to solve the above-mentioned problems, one embodiment of the present invention provides a method for producing a molded article containing a polyhydroxyalkanoate resin, which includes a step of producing the molded article by blending at least two or more raw materials extracted from microorganisms using process values ​​calculated by a simulation that simulates an extruder used to produce the molded article.

[0012] The information processing device according to each aspect of the present invention may be realized by a computer. In this case, an information processing program that causes the computer to operate as each unit (software element) of the information processing device to realize the information processing device on the computer, and a computer-readable recording medium on which the program is recorded, also fall within the scope of the present invention. In addition, such an information processing program may be stored in a cloud server and provided so as to be downloadable from the cloud server. [Effects of the Invention]

[0013] According to one aspect of the present invention, it is possible to accurately predict the quality of a molded product. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a diagram illustrating an overview of an information processing system according to a first embodiment of the present invention. [Figure 2]1 is a block diagram showing an example of a configuration of a main part of an information processing device according to a first embodiment of the present invention. [Figure 3] 1 is a graph showing the correlation between resin temperature and MFR. [Figure 4] 1 is a graph showing the correlation of resin temperature and MFR with respect to discharge amount. [Figure 5] 10 is a graph showing an example of a simulation result using a simulation program. [Figure 6] FIG. 2 is a schematic diagram showing an example of the flow of each piece of data according to the first embodiment of the present invention. [Figure 7] 4 is a flowchart showing an example of processing executed by the information processing device according to the first embodiment of the present invention. [Figure 8] FIG. 10 is a block diagram showing an example of a configuration of a main part of an information processing device according to a second embodiment of the present invention. [Figure 9] FIG. 10 is a schematic diagram showing an example of the flow of each piece of data according to the second embodiment of the present invention. [Figure 10] 10 is a flowchart showing an example of processing executed by an information processing device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] [Embodiment 1] Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. In the description of the drawings, the same parts are given the same reference numerals and the description will be omitted.

[0016] (Outline of information processing system) An overview of an information processing system 100 according to this embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram showing an overview of the information processing system 100. The information processing system 100 is a system that can be used to support extrusion molding, and includes an extruder 90, an information processing device 1, and an information processing device 2. Below, the extrusion molding to be supported will be described, followed by a description of the information processing device 1 and the information processing device 2. Note that the information processing system 100 is not limited to supporting extrusion molding, and may also be used to support injection molding.

[0017] The extruder 90 includes a hopper 92, a side feeder 93, a barrel 94, a screw 91 housed inside the barrel 94, and a die head 95. The extruder 90 may be a single-screw extruder equipped with one screw, or a twin-screw extruder equipped with two screws. Hereinafter, unless otherwise specified, the extruder 90 will be described as a twin-screw extruder. Furthermore, when it is not necessary to distinguish between the two screws, the two screws will be collectively described as screw 91.

[0018] The barrel 94 includes a cylinder (not shown) having a shape of two combined cylinders, and a heater (not shown) for melting the raw material. The raw material is fed from a hopper 92 provided at the top of the barrel 94 and supplied to the cylinder. The raw material fed from the hopper 92 is melted by the heat of the heater.

[0019] The screws 91 are housed in a cylinder in a state where they mesh with each other, and rotate in the same direction when they receive torque from a motor (not shown). This allows the raw materials to be kneaded. In addition, predetermined additives and the like are fed from a side feeder 93, and the additives are kneaded with the raw materials.

[0020] The raw materials and additives thus fed into the extruder 90 are melted and kneaded, processed into a specific shape (for example, pellets), and extruded from the die head 95. Note that a description of the cooling process and the like after extrusion from the die head 95 will be omitted.

[0021] An example of a resin molded product produced by the extruder 90 is pellets used in the manufacture of plastic products. Pellets are formed by blending multiple raw materials, and the raw materials in this embodiment include polymers. The polymer is not particularly limited, but may be the biodegradable polymer described above, for example. In the following, unless otherwise specified, the polymer contained in the raw material will be described as a biodegradable polymer. However, with regard to the blended polymers, all of the polymers may be biodegradable polymers, or only some of the polymers may be biodegradable polymers.

[0022] In the manufacturing process using the extruder 90, polymers extracted from microorganisms are melted and kneaded with predetermined additives, and then processed into pellets. Polymers are classified into predetermined types based on their molecular weight structure and functional group composition, and pellets are usually produced by blending two to four types of polymers. The melting characteristics of the pellets and their physical properties after molding can be manipulated by changing the type and blending ratio of the polymers blended. The pellets produced are classified into predetermined grades depending on their intended use. For each grade, a standard recipe is established that determines the type and blending ratio of polymers to reproduce quality.

[0023] As mentioned above, the production process of biodegradable polymers is based on biochemical processes, and therefore, differences in physical properties may occur between culture lots of the same type of polymer. Therefore, even if the polymers are blended according to a standard recipe, the desired quality may not be achieved, i.e., the quality may deviate from the specifications. Therefore, in this embodiment, a quality prediction model is used to predict the quality of pellets (resin molded products).

[0024] The information processing device 1 generates a quality prediction model through machine learning using training data, and the information processing device 2 predicts the quality of a resin molded product from prediction data using the quality prediction model.

[0025] (Configuration examples of information processing device 1 and information processing device 2) FIG. 2 is a block diagram showing an example of the configuration of the information processing device 1 and the information processing device 2. As shown in FIG.

[0026] 2, the information processing device 1 includes a control unit 10 that controls each unit of the information processing device 1, and a storage unit 11 that stores various data used by the information processing device 1. The information processing device 1 also includes a communication unit 12 that enables the information processing device 1 to communicate with other devices, an input unit 13 that accepts input of various data to the information processing device 1, and an output unit 14 that enables the information processing device 1 to output various data. The control unit 10 also includes a training data acquisition unit 101 and a learning unit 102. The storage unit 11 stores training data 111.

[0027] The information processing device 1 may be configured as a general-purpose computer. When the information processing device 1 is configured as a general-purpose computer, the control unit 10 may be configured to include a processor, a read-only memory (ROM), a random access memory (RAM), and the like. The processor can read programs from the ROM and execute various programs using the RAM as a work area. The storage unit 11 may be configured as a hard disk drive (HDD) or a solid state drive (SSD). The storage unit 11 does not necessarily have to be provided in the information processing device 1. In this case, for example, the storage unit 11 may be provided in an external server. Such an external server may be a virtual server running on a hypervisor or a cloud server. When the storage unit 11 is provided in an external server, the information processing device 1 can access the storage unit 11 via a communication network such as a wireless local area network (LAN). As described above, this embodiment includes both a case where the information processing device 1 is physically connected to the storage unit 11 and a case where the information processing device 1 is not physically connected to the storage unit 11.

[0028] The input unit 13 is typically a keyboard or a mouse, but is not limited to these, and may be any configuration that allows various data to be input. For example, the input unit 13 may be a microphone, a touch panel, etc. The output unit 14 is typically a display, but is not limited to these, and may be any configuration that allows various data to be output. For example, the output unit 14 may be a speaker, a warning light, etc.

[0029] The training data 111 stored in the storage unit 11 will be described. The training data 111 is labeled data used in machine learning. Machine learning is a technique for learning features contained in input data and generating a "model" that predicts results corresponding to newly input data. A quality prediction model 211, which will be described later, is a trained model generated through machine learning.

[0030] The training data 111 is generated by associating input data with correct answer data, and is stored in the storage unit 11. An example of the "input data" here is the physical property conditions that indicate the physical properties of a resin molded product (e.g., pellets) manufactured by blending multiple raw materials (polymers), and the process values ​​during processing of the raw materials calculated by a simulation that simulates the extruder 90 used to manufacture the resin molded product.

[0031] Here, the "physical property conditions" in this embodiment will be described. An example of a physical property condition is MFR potential (Melt Flow Rate), which indicates the physical properties of a resin molded product. MFR is a well-known physical property index for evaluating the fluidity of a resin. The definition of MFR potential is not particularly limited, but may be defined as an MFR that is not affected by changes or fluctuations in production conditions, such as when all samples are manufactured under the same conditions.

[0032] Next, the "process values" in this embodiment will be described. Generally, process values ​​are explained as various actual values ​​measured during production. Taking the extruder 90 as an example, typical process values ​​are the resin temperature and resin pressure during the production of a resin molded product. These actual values ​​are measured by temperature sensors, pressure sensors, etc. installed in the extruder 90.

[0033] In contrast, the process values ​​in this embodiment are not actual values ​​measured by temperature sensors, pressure sensors, etc., but virtual values ​​calculated by a simulation that simulates the extruder 90. Unless otherwise specified, the process values ​​simply referred to as process values ​​hereinafter refer to virtual values ​​calculated by a simulation that simulates the extruder 90. Such process values ​​include, for example, the barrel temperature and resin pressure in each manufacturing process.

[0034] Next, an example of "correct answer data" will be described. Correct answer data is data that uses "physical property conditions" and "process values" as input data and indicates the quality of a resin molded product manufactured in accordance with these physical property conditions and process values. Hereinafter, "data indicating the quality of a resin molded product manufactured in accordance with the physical property conditions and process values" may be referred to as quality data.

[0035] The training data 111 is generated by associating the above-mentioned input data with the correct answer data, and is stored in the storage unit 11.

[0036] The training data acquisition unit 101 refers to the storage unit 11 to acquire training data 111. The training data acquisition unit 101 outputs the acquired training data 111 to the learning unit .

[0037] The learning unit 102 generates a quality prediction model 211 for predicting the quality of a resin molded product from the prediction data through machine learning using the training data 111 acquired from the training data acquisition unit 101. The quality prediction model 211 is a computational model that uses physical property conditions and process values ​​as explanatory variables and the quality of the resin molded product as a target variable. The machine learning algorithm is not particularly limited. For example, the learning unit 102 may generate the quality prediction model 211 using a neural network, or may generate the quality prediction model 211 using regression analysis, random forest, or the like.

[0038] The quality prediction model 211 generated by the learning unit 102 is transmitted to the information processing device 2 and stored in the storage unit 21 of the information processing device 2. The quality prediction model 211 generated by the learning unit 102 may be stored in the storage unit 11.

[0039] Next, the information processing device 2 will be described. The information processing device 2 includes a control unit 20 that controls each unit of the information processing device 2 in an integrated manner, and a storage unit 21 that stores various data used by the information processing device 2. The information processing device 2 also includes a communication unit 22 that enables the information processing device 2 to communicate with other devices, an input unit 23 that accepts various data input to the information processing device 2, and an output unit 24 that enables the information processing device 2 to output various data. The control unit 20 also includes a condition acquisition unit 201, a quality prediction unit 202, and an output control unit 203. The storage unit 21 stores a quality prediction model 211 and a simulation program 212. The simulation program 212 is a program that simulates the extruder 90, and is capable of simulating a manufacturing process using the extruder 90.

[0040] The information processing device 2 may also be configured as a general-purpose computer, similar to the information processing device 1. When the information processing device 2 is configured as a general-purpose computer, the control unit 20, the storage unit 21, the communication unit 22, the input unit 23, and the output unit 24 may have the same configurations as the control unit 10, the storage unit 11, the communication unit 12, the input unit 13, and the output unit 14 of the information processing device 1. In other words, the storage unit 21 may also be provided in an external server, rather than in the information processing device 2. This embodiment includes both a case where the information processing device 2 is physically connected to the storage unit 21 and a case where the information processing device 2 is not physically connected to the storage unit 21.

[0041] The condition acquisition unit 201 acquires physical property conditions and process values. A method for acquiring physical property conditions will be described later. Here, an example of a method for acquiring process values ​​will be described. The condition acquisition unit 201 can acquire process values ​​such as barrel temperature and resin pressure in each manufacturing process by using a simulation program 212. The condition acquisition unit 201 outputs the acquired physical property conditions and process values ​​to the quality prediction unit 202. The physical property conditions and process values ​​acquired by the condition acquisition unit 201 can be said to be prediction data for predicting the quality of a resin molded product.

[0042] In this embodiment, the process values ​​calculated by the simulation simulating the extruder 90 do not include the resin temperature (virtual resin temperature) measured near the tip of the extruder 90. The reason for not including the resin temperature in the process values ​​is that, as described above, the resin temperature measured only near the tip of the extruder 90 does not adequately reflect the effects of the production conditions. This point will be explained in detail using FIGS. 3 and 4.

[0043] Figure 3 is a graph showing the correlation between resin temperature (°C) and MFR (g / 10 min). In Figure 3, the horizontal axis represents resin temperature, and the vertical axis represents MFR. Figure 3 shows the correlation between resin temperature and MFR when production conditions do not change. As is clear from Figure 3, if the production conditions are the same, there is a strong correlation between resin temperature and MFR, and it can be seen that as the resin temperature increases, the MFR also increases.

[0044] Fig. 4 is a graph showing the correlation between the resin temperature (°C) and the MFR (g / 10 min) versus the discharge rate (kg / h) of the extruder 90. In Fig. 4, the horizontal axis represents the discharge rate of the extruder 90, the right vertical axis represents the resin temperature, and the left vertical axis represents the MFR.

[0045] As shown in FIG. 4, when the discharge rate, which is one of the production conditions, fluctuates, the correlation between the resin temperature and the MFR becomes reversed compared to that shown in FIG. 3. Specifically, as the discharge rate increases, the resin temperature rises while the MFR decreases. This is the opposite of the correlation shown in FIG. 3. Thus, FIG. 4 suggests that the resin temperature measured near the tip of the extruder 90, which is an explanatory variable affecting the quality of a resin molded product, may not achieve the desired quality. Therefore, in this embodiment, the resin temperature measured near the tip of the extruder 90 is excluded from the production conditions included in the process value, and barrel temperature, resin pressure, and other factors in each manufacturing process are included in the production conditions instead of the resin temperature measured near the tip of the extruder 90. Details of the production conditions included in the process value will be described later. Note that the resin temperature may be included in the process value excluding the resin temperature measured near the tip of the extruder 90. This point will also be described later.

[0046] 2 , the quality prediction unit 202 predicts the quality of the resin molded product from the physical property conditions and process values, which are prediction data, using a quality prediction model 211 generated by learning the relationship between the physical property conditions and process values ​​and the quality data of the resin molded product. Specifically, the quality prediction unit 202 can predict the quality of the resin molded product by inputting the physical property conditions and process values ​​acquired by the condition acquisition unit 201 into the quality prediction model 211. The quality of the resin molded product predicted by the quality prediction unit 202 is output to the output control unit 203.

[0047] The output control unit 203 causes the output unit 24 to output the prediction result by the quality prediction unit 202. The output mode is arbitrary, and for example, the output control unit 203 may cause the prediction result to be output in at least one of the following modes: display output, audio output, and print output. Furthermore, the device that outputs the prediction result may be an external device to the information processing device 2.

[0048] Furthermore, the output control unit 203 may output information according to the prediction result by the quality prediction unit 202. For example, when the prediction result by the quality prediction unit 202 indicates a sign of deviation from specifications, the output control unit 203 may output a notice urging correction of the blending ratio of raw materials.

[0049] As described above, the information processing device 2 includes a condition acquisition unit 201 that acquires physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that mimics the extruder 90 used to manufacture the molded product, and a quality prediction unit 202 that predicts the quality of the molded product from the physical property conditions and process values ​​using a quality prediction model 211 that uses the physical property conditions and process values ​​acquired by the condition acquisition unit 201 as explanatory variables and the quality of the molded product as a target variable.

[0050] Fig. 5 is a graph showing an example of the results of a simulation using the simulation program 212. In Fig. 5, the horizontal axis represents the axial distance of the extruder 90, the right vertical axis represents the resin temperature, and the left vertical axis represents the volume, resin pressure, and residence time.

[0051] Figure 5 shows seven production conditions calculated by a simulation simulating an extruder 90 among the production conditions when manufacturing a resin molded product. The seven production conditions are indicated by reference numerals 80 to 86. Reference numeral 80 indicates the resin filling rate in the space surrounded by the barrel 94 and the screw elements. Reference numeral 81 indicates the volume of the space surrounded by the barrel 94 and the screw elements. Reference numeral 82 indicates the residence time, which is the time the raw material remains inside the extruder 90. Reference numeral 83 indicates the integrated value of the residence time. Reference numeral 84 indicates the resin pressure. Reference numeral 85 indicates the resin temperature. Reference numeral 86 indicates the barrel temperature.

[0052] In FIG. 5, the area around 100 mm indicates the area where raw materials are introduced from the hopper 92. The area from around 100 mm to around 2000 mm indicates the area where the raw materials are melted by the heat of the heater. The areas around 2100 mm and 3100 mm indicate the areas where the raw materials are kneaded. The area around 2600 mm indicates the area where additives are introduced from the side feeder 93. The area around 3800 mm indicates the area where the resin is discharged from the die head 95.

[0053] The resin filling rate increases in the kneading region, as indicated by reference numeral 80. In the region where the resin filling rate increases, the volume of the space decreases, as indicated by reference numeral 81. Furthermore, as indicated by reference numeral 83, the residence time gradually increases toward the die head 95 of the extruder 90.

[0054] Furthermore, resin pressure rises in the kneading and discharging regions as indicated by reference numeral 84. Furthermore, the resin temperature rises in the kneading and discharging regions due to heating by the heater and shear heat generated by kneading as indicated by reference numeral 85. Furthermore, the barrel temperature is controlled by the heater as indicated by reference numeral 86, and differs for each manufacturing process (e.g., melting process, kneading process, discharging process, etc.).

[0055] The inventors noticed that it is possible to calculate the amount of heat received by the resin in each manufacturing process from the results of such simulations. Specifically, the amount of heat received by the resin in each manufacturing process can be calculated from the barrel temperature in each manufacturing process, the residence time in each manufacturing process, and the resin pressure in each manufacturing process. Note that since the method for calculating the amount of heat is well known, its explanation will be omitted.

[0056] It is known that the molecular weight of the resin being kneaded in the extruder 90 decreases due to heating by the heater and shear heat caused by kneading, causing the MFR to fluctuate. In response to this, by using simulation, it is possible to calculate the amount of heat received by the resin from the barrel temperature, residence time, and resin pressure, even when the production conditions change. Furthermore, by using at least the barrel temperature and residence time as explanatory variables of the quality prediction model 211 and inputting these data obtained from the simulation into the quality prediction model 211, it becomes possible to predict the MFR more accurately. Furthermore, the explanatory variables of the quality prediction model 211 may further include the resin pressure and / or the amount of heat. This further improves the accuracy of MFR prediction.

[0057] Although the heat quantity received by the resin is calculated from the barrel temperature, residence time, and resin pressure as described above, the calculation method is not limited to this. For example, the heat quantity received by the resin may be calculated from only the barrel temperature.

[0058] As described above, the process values ​​calculated by the simulation during processing of the raw materials include at least the temperature of the extruder 90 during a predetermined manufacturing process and the residence time of the raw materials inside the extruder 90. Preferably, the process values ​​may further include the pressure of the raw materials and / or the calorific value of the raw materials calculated from at least the temperature of the extruder 90. The "predetermined manufacturing process" here includes at least the melting process, kneading process, and discharge process described above. The "temperature of the extruder 90" may also be referred to as the barrel temperature described above. Since the barrel 94 is equipped with a heater, the barrel temperature may also be referred to as the temperature of the heater provided in the barrel 94. Since the heater provided in the barrel 94 is used to melt the raw materials as described above, the barrel temperature may also be referred to as the resin temperature. In other words, the "temperature of the extruder 90" may also be referred to as the resin temperature. However, as described above, the resin temperature measured near the tip of the extruder 90 is not included in the process values. That is, the resin temperature in the melting step and / or the kneading step may be included in the process value, but the resin temperature in the discharging step is not included in the process value.

[0059] (Data flow during manufacturing) FIG. 6 is a schematic diagram showing an example of the flow of data during the manufacture of a resin molded product.

[0060] When a resin molded product is being manufactured, the condition acquisition unit 201 acquires the physical property conditions 40 and the process values ​​41. The physical property conditions 40 and the process values ​​41 acquired by the condition acquisition unit 201 are input to the quality prediction model 211.

[0061] The quality prediction model 211 predicts the quality of a resin molded product based on the physical property conditions 40 and the process values ​​41. The prediction result is output as, for example, an MFR. The physical property conditions 40 and the process values ​​41 are explanatory variables in the quality prediction model 211, and the quality of the resin molded product is a target variable in the quality prediction model 211.

[0062] (Processing flow) Next, the flow of processing executed by the information processing device 2 will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of processing executed by the information processing device 2.

[0063] In step S101, the condition acquisition unit 201 acquires prediction data to be input to the quality prediction model 211. The prediction data includes at least physical property conditions and process values. The condition acquisition unit 201 acquires the prediction data, for example, when an operator is manufacturing a resin molded product.

[0064] The process proceeds to step S102, where the quality prediction unit 202 inputs the prediction data acquired in the process of step S101 into the quality prediction model 211 to predict the quality of the resin molded product.

[0065] The process proceeds to step S103, where the output control unit 203 outputs the quality of the resin molded product obtained in the process of step S102 to the output unit 24. This allows, for example, an operator to grasp the quality of the resin molded product that has been accurately predicted.

[0066] As described above, the information processing method according to the first embodiment includes a condition acquisition step (S101) for acquiring physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that mimics an extruder used to manufacture the molded product, and a quality prediction step (S102) for predicting the quality of the molded product from the physical property conditions and process values ​​using a quality prediction model that uses the physical property conditions and process values ​​as explanatory variables and the quality of the molded product as a target variable.

[0067] (Action and effect) As described above, according to the first embodiment, the following advantageous effects can be obtained.

[0068] In the first embodiment, a quality prediction model 211 is used, in which the physical property conditions and the process values ​​calculated by a simulation simulating the extruder 90 are used as explanatory variables, to predict the quality of a resin molded product (e.g., pellets) from the physical property conditions and the process values.

[0069] The process values ​​calculated by the simulation include at least the barrel temperature in each manufacturing process and the residence time in each manufacturing process. It is known that the molecular weight of the resin being kneaded in the extruder 90 decreases due to heating by the heater and shear heat caused by kneading, causing fluctuations in the MFR. In contrast, by using the process values ​​obtained by the simulation as described above, it becomes possible to accurately predict the quality (MFR) of the resin molded product, even when production conditions change. Furthermore, the process values ​​calculated by the simulation may further include the resin pressure in each manufacturing process and / or the amount of heat received by the resin, calculated from at least the barrel temperature. This further improves the accuracy of MFR prediction.

[0070] It is conceivable to use the resin temperature measured near the tip of the extruder 90 as an explanatory variable of the quality prediction model 211, but the resin temperature measured near the tip of the extruder 90 does not sufficiently reflect the effects of the production conditions, and there is a concern that the measurement accuracy may decrease. According to the above configuration, the explanatory variables of the quality prediction model 211 do not include the resin temperature measured near the tip of the extruder 90, so that the explanatory variables can be made to more accurately reflect the production conditions, and by inputting such explanatory variables, it becomes possible to accurately predict the quality of the resin molded product.

[0071] Furthermore, the prediction results of the quality prediction model 211 may be used to optimize the production conditions of resin molded products (for example, the blending ratio of each polymer). A well-known optimization method may be used for the optimization calculation. Well-known optimization methods include grid search, random search, genetic algorithm, and Bayesian optimization. The optimized production conditions may be output to the output unit 24. This allows the operator to understand the optimized production conditions, making it possible to manufacture products with the desired quality. Furthermore, if the information processing device 2 is configured with a PLC (Programmable Logic Controller), it is possible to automatically apply the optimized production conditions, making it possible to automatically manufacture products with the desired quality.

[0072] Furthermore, with the above configuration, it is possible to calculate the amount of heat the resin will receive even after changing the blending ratio of the raw materials, which is expected to improve the accuracy when adjusting the MFR so that the quality does not deviate from the standard.

[0073] [Embodiment 2] FIG. 8 is a block diagram showing an example of a main configuration of an information processing device 3 according to the second embodiment of the present invention.

[0074] 8, the information processing device 3 includes a control unit 30 that controls each unit of the information processing device 3, and a storage unit 31 that stores various data used by the information processing device 3. The information processing device 3 also includes a communication unit 32 that enables the information processing device 3 to communicate with other devices, an input unit 33 that accepts various data input to the information processing device 3, and an output unit 34 that enables the information processing device 3 to output various data. The control unit 30 also includes a physical property condition prediction unit 301, a condition acquisition unit 302, a quality prediction unit 303, and an output control unit 304. The storage unit 31 stores a quality prediction model 311, a simulation program 312, and a physical property condition prediction model 313.

[0075] The functions of the condition acquisition unit 302, quality prediction unit 303, output control unit 304, communication unit 32, input unit 33, and output unit 34 included in the information processing device 3 are similar to the functions of the condition acquisition unit 201, quality prediction unit 202, output control unit 203, communication unit 22, input unit 23, and output unit 24 included in the information processing device 2 according to the first embodiment, and therefore descriptions thereof will be omitted where appropriate. Furthermore, the quality prediction model 311 and simulation program 312 stored in the storage unit 31 are similar to the quality prediction model 211 and simulation program 212 stored in the storage unit 21 according to the first embodiment, and therefore descriptions thereof will be omitted where appropriate.

[0076] The physical property condition prediction model 313 stored in the memory unit 31 uses as input data blending ratio conditions indicating the blending ratio of multiple polymers and physical property values ​​indicating the physical properties of multiple polymers, and predicts the physical property conditions of a resin molded product to be manufactured when these conditions are applied.

[0077] Like the quality prediction model 311, the physical property condition prediction model 313 is also a model generated by machine learning. However, when generating the physical property condition prediction model 313, training data (data in which input data is associated with correct answer data) used for machine learning is different from the training data used for machine learning of the quality prediction model 311. Specifically, when generating the physical property condition prediction model 313, training data different from the training data 111 according to the first embodiment is used. Hereinafter, the training data different from the training data 111 according to the first embodiment will be referred to as training data A. The training data A will be described below.

[0078] Training data A is a correspondence between "input data (which can also be called explanatory variables)" and "correct answer data (which can also be called target variables)." This "input data" includes various data related to the physical property conditions to be predicted. Specifically, the "input data" includes the above-mentioned blending ratio conditions and physical property values.

[0079] Here, an example of "physical property values" will be described. Physical property values ​​may include molecular weight data, thermal stability, thermal decomposition temperature, MFR, etc. of each polymer to be blended. Molecular weight data may include number average molecular weight (Mn), weight average molecular weight (Mw), molecular weight distribution (Mw / Mn), etc. "Thermal stability" here refers to the degree of molecular weight reduction caused by treatment under a specified temperature and pressure for a specified time, and the unit used is %.

[0080] Furthermore, the "physical properties" may include the content of repeating units of a predetermined functional group possessed by each polymer to be blended. The "physical properties" need only include at least one of the above data. Of course, the "physical properties" may include all of the above data, or may include data other than the above data.

[0081] The "correct answer data" in training data A is data that indicates the quality of a resin molded product manufactured in accordance with "mixing ratio conditions" and "physical property values" as input data.

[0082] In the second embodiment, the "quality of the resin molded product" includes two types of quality. One is the "quality of the resin molded product manufactured in accordance with the blending ratio conditions and physical property values" described above, and the other is the "quality of the resin molded product manufactured in accordance with the physical property conditions and process values" described in the first embodiment.

[0083] The "quality of a resin molded product manufactured in accordance with the blending ratio conditions and physical property values" is also referred to as a physical property condition. An example of this physical property condition is the MFR potential, as described in embodiment 1. Also, an example of the "quality of a resin molded product manufactured in accordance with the physical property conditions and process values" is the MFR, as described in embodiment 1.

[0084] Therefore, "the quality of a resin molded product manufactured in accordance with physical property conditions and process values" may be rephrased as "the MFR of a resin molded product manufactured in accordance with the MFR potential and process values."

[0085] The MFR potential may be defined as described in embodiment 1, or may be defined as data indicating the quality of a resin molded product manufactured in accordance with the blending ratio conditions and physical property values. As described in embodiment 1, when the MFR potential is defined as an MFR that is not affected by changes or fluctuations in production conditions, such as when all samples are manufactured under the same conditions, the relative differences in MFR potential between samples can be explained only by the blending ratio conditions and physical property values.

[0086] A data set generated by associating the input data with the correct answer data becomes training data A for generating the physical property condition prediction model 313. The process of generating the training data A may be performed by the information processing device 1 according to the first embodiment or by another device.

[0087] The physical property condition prediction unit 301 acquires input data to be input to the physical property condition prediction model 313. The input data includes blending ratio conditions and physical property values. There is no particular limitation on the method of acquiring the blending ratio conditions and physical property values. For example, the physical property condition prediction unit 301 may acquire data entered into the input unit 33 by an operator engaged in the manufacture of resin molded products as the blending ratio conditions and physical property values. Furthermore, if the information processing device 3 is configured as a PLC, the physical property condition prediction unit 301 may acquire the blending ratio conditions and physical property values ​​set in the PLC. The blending ratio conditions and physical property values ​​acquired by the physical property condition prediction unit 301 can be said to be data for predicting physical property conditions.

[0088] The physical property condition prediction unit 301 predicts the physical property conditions from the blending ratio conditions and physical property values ​​using a physical property condition prediction model 313 generated by learning the relationship between the blending ratio conditions and physical property values ​​and the physical property conditions of the resin molded product. Specifically, the physical property condition prediction unit 301 can predict the physical property conditions by inputting the blending ratio conditions and physical property values ​​to the physical property condition prediction model 313. The physical property conditions predicted by the physical property condition prediction unit 301 are output to the condition acquisition unit 302.

[0089] The condition acquisition unit 302 acquires the physical property conditions of the resin molded product predicted by the physical property condition prediction unit 301. The condition acquisition unit 302 also acquires process values ​​using a simulation program 312. The physical property conditions and process values ​​acquired by the condition acquisition unit 302 are output to the quality prediction unit 303.

[0090] The quality prediction unit 303 predicts the quality of the resin molded product from the physical property conditions and process values ​​using a quality prediction model 311 generated by learning the relationship between the physical property conditions and process values ​​and the quality data of the resin molded product. The quality of the resin molded product predicted by the quality prediction unit 303 is output to the output control unit 304.

[0091] As described above, if the "quality of a resin molded product manufactured in accordance with physical property conditions and process values" is rephrased as the "MFR of a resin molded product manufactured in accordance with the MFR potential and process values," then "predicting the MFR of a resin molded product" may also mean "correcting the MFR potential of a resin molded product." In this case, the MFR potential of a resin molded product may be rephrased as data indicating the quality of the resin molded product before correction, and the MFR of a resin molded product may be rephrased as data indicating the quality of the resin molded product after correction.

[0092] The output control unit 304 causes the output unit 34 to output the quality of the resin molded product predicted by the quality prediction unit 303. The output may be in any form, and for example, the output control unit 304 may output the predicted quality of the resin molded product in at least one form of display output, audio output, or print output. Furthermore, the device that outputs the predicted quality of the resin molded product may be a device external to the information processing device 3. Such an output allows the operator to grasp the accurately predicted quality of the resin molded product.

[0093] As described above, the information processing device 3 includes a physical property condition prediction unit 301 that predicts physical property conditions based on the blending ratio conditions indicating the ratio at which polymers are blended and at least one physical property value selected from the molecular weight data of the polymers, the thermal stability of the polymers, the thermal decomposition temperature of the polymers, the melt flow rate of the polymers, and the content of repeating units of a predetermined functional group possessed by the polymers.

[0094] (Data flow during manufacturing) Fig. 9 is a schematic diagram showing an example of the flow of each piece of data during the manufacture of a resin molded product. In the flow of Fig. 9, a flow for predicting physical property conditions is added to the flow of Fig. 6. The description of the flow explained in Fig. 6 will be omitted where appropriate.

[0095] 9 may be conditions indicating the blending ratio of multiple polymers specified in a standard recipe for each grade. The physical property values ​​51 may be data indicating the physical properties of each polymer to be blended. The blending ratio conditions 50 and the physical property values ​​51 are acquired by the physical property condition prediction unit 301 and input to the physical property condition prediction model 313.

[0096] The physical property condition prediction model 313 predicts the physical property conditions of the resin molded product based on the inputted blending ratio conditions 50 and physical property values ​​51. The blending ratio conditions 50 and the physical property values ​​51 are explanatory variables in the physical property condition prediction model 313, and the physical property conditions of the resin molded product are objective variables in the physical property condition prediction model 313. The physical property conditions output from the physical property condition prediction model 313 are input to the quality prediction model 311 together with the process values ​​41.

[0097] The blending ratio conditions 50 and physical property values ​​51 used to generate the physical property condition prediction model 313 and the physical property conditions 40 (see FIG. 6) and process values ​​41 used to generate the quality prediction model 311 may be acquired in different facilities. For example, the blending ratio conditions 50 and physical property values ​​51 may be acquired in research facilities, while the physical property conditions 40 and process values ​​41 may be acquired in manufacturing facilities. This makes it possible to acquire data for model generation without affecting the production plan for resin molded products (e.g., pellets). Of course, the above data may be acquired in the same facility (e.g., manufacturing facility) as long as it satisfies certain conditions, such as not interfering with the pellet production plan.

[0098] (Processing flow) Next, the flow of processing executed by the information processing device 3 will be described with reference to Fig. 10. Fig. 10 is a flowchart showing an example of processing executed by the information processing device 3. Note that the processing shown in steps S202, S203, and S204 is similar to the processing shown in steps S101, S102, and S103 shown in Fig. 7, and therefore description thereof will be omitted where appropriate.

[0099] In step S201, the physical property condition prediction unit 301 acquires input data to be input to the physical property condition prediction model 313. The input data includes blending ratio conditions and physical property values. The timing when the physical property condition prediction unit 301 acquires the blending ratio conditions and physical property values ​​is, for example, when an operator is manufacturing a resin molded product.

[0100] The physical property condition prediction unit 301 predicts the physical property conditions from the blending ratio conditions and physical property values ​​using a physical property condition prediction model 313 generated by learning the relationship between the blending ratio conditions and physical property values ​​and the physical property conditions of the resin molded product.

[0101] The process proceeds to step S202, where the condition acquisition unit 302 acquires prediction data to be input to the quality prediction model 311. The prediction data includes at least the process values ​​and the physical property conditions predicted in the process of step S201.

[0102] The process proceeds to step S203, where the quality prediction unit 303 inputs the prediction data acquired in the process of step S202, ie, the physical property conditions and process values, into the quality prediction model 311 to predict the quality of the resin molded product.

[0103] The process proceeds to step S204, where the output control unit 304 outputs the quality of the resin molded product predicted in the process of step S203 to the output unit 34. This allows, for example, an operator to grasp the quality of the resin molded product that has been predicted with high accuracy.

[0104] (Action and effect) As described above, according to the second embodiment, the following advantageous effects can be obtained.

[0105] In the second embodiment, the physical property conditions of the resin molded product are predicted from the polymer blending ratio conditions and the physical property values ​​of the polymer, and the predicted physical property conditions are input to the quality prediction model 311.

[0106] According to the above configuration, by using the polymer blending ratio conditions and the polymer physical property values ​​to accurately predict the physical property conditions as explanatory variables of the quality prediction model 311, it becomes possible to accurately predict the quality of the resin molded product.

[0107] Furthermore, in the second embodiment, a two-stage prediction process is adopted in which output from the physical property condition prediction model 313 is input to the quality prediction model 311. In machine learning, a large number of variables are required to generate a computational model that performs quality prediction as in the second embodiment, but by adopting such a two-stage prediction process, fewer variables are required for each model, contributing to a reduction in the amount of required data. Furthermore, generating models with fewer variables also contributes to improved flexibility and adaptability in model management.

[0108] Furthermore, the molded article described in each embodiment may be a molded article containing a polyhydroxyalkanoate resin, which is a biodegradable resin. The manufacturing method of the molded article described in each embodiment can also be expressed as follows. That is, the manufacturing method of a molded article containing a polyhydroxyalkanoate resin may include a step of manufacturing the molded article by blending at least two or more raw materials extracted from microorganisms using process values ​​calculated by a simulation simulating an extruder 90 used in manufacturing the molded article.

[0109] [Software implementation example] The functions of the information processing devices 1 to 3 (hereinafter simply referred to as "devices") can be realized by a program (information processing program) for causing a computer to function as the device, and a program for causing a computer to function as each control block of the device (particularly each part included in control unit 10, control unit 20, and control unit 30).

[0110] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The computer executes the program to realize each function described in each embodiment.

[0111] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0112] In addition, some or all of the functions of each control block can be realized by a logic circuit. For example, an integrated circuit in which a logic circuit that functions as each control block is formed is also included in the scope of the present invention. In addition, the functions of each control block can also be realized by, for example, a quantum computer.

[0113] Furthermore, each process described in each embodiment may be executed by AI (Artificial Intelligence). In this case, the AI ​​may run on the above-mentioned device or on another device (for example, an edge computer or a cloud server).

[0114] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0115] 1, 2, 3 Information processing equipment 40 Physical property conditions 41 Process Values 50 Mixing ratio conditions 51 Physical properties 90 Extruder 201, 302 Condition acquisition section 202, 303 Quality Prediction Department 211, 311 Quality prediction model 212, 312 Simulation Program 301 Physical Property Condition Prediction Department

Claims

1. a condition acquisition unit that acquires physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that simulates an extruder used to manufacture the molded product; a quality prediction unit that predicts the quality of the molded product from the physical property conditions and the process values ​​acquired by the condition acquisition unit using a quality prediction model that uses the physical property conditions and the process values ​​acquired by the condition acquisition unit as explanatory variables and the quality of the molded product as a target variable; Information processing device.

2. The process values ​​during processing of the raw material calculated by the simulation include at least the temperature of the extruder in a predetermined manufacturing process and the residence time of the raw material inside the extruder. The information processing device according to claim 1 .

3. the raw material is a polymer, a physical property condition prediction unit that predicts the physical property conditions based on a blending ratio condition indicating a blending ratio of the polymers and at least one physical property value selected from molecular weight data of the polymers, thermal stability of the polymers, thermal decomposition temperature of the polymers, melt flow rate of the polymers, and a content of repeating units of a predetermined functional group contained in the polymers; 3. The information processing device according to claim 1 or 2.

4. An information processing method executed by one or more information processing devices, a condition acquisition step of acquiring physical property conditions that indicate the physical properties of a molded product manufactured by blending at least two or more raw materials extracted from microorganisms, and process values ​​during processing of the raw materials that are calculated by a simulation that simulates an extruder used to manufacture the molded product; a quality prediction step of predicting the quality of the molded product from the physical property conditions and the process values ​​using a quality prediction model that uses the physical property conditions and the process values ​​as explanatory variables and the quality of the molded product as a target variable, Information processing methods.

5. 2. An information processing program for causing a computer to function as the information processing device according to claim 1, the information processing program causing a computer to function as the condition acquisition unit and the quality prediction unit.

6. A method for producing a molded article containing a polyhydroxyalkanoate resin, comprising: The method includes a step of blending at least two or more raw materials extracted from microorganisms to manufacture the molded product using process values ​​calculated by a simulation simulating an extruder used to manufacture the molded product. Manufacturing method of molded products.

Images