Blow molding machine operating condition inference device, method and program, and learning device

Abstract

To assist various settings of a blow molding machine.SOLUTION: A learned model 73 is prepared using the physical property information of a preform, the molding conditions of bottles, the measured values of the physical property information of the bottles, and the measured environmental information at the time of manufacturing the bottles as input data, and the operating conditions of the blow molding machine as output data. By inputting the target values 81a of the physical property information to be satisfied by the bottles after molding, as well as the measured values 81b of the physical property information of the preform, the molding conditions of bottles, and the measured environmental information into the learned model 73, the operating conditions (heater output, blow pressure, blow timing, rod speed) of the blow molding machine to be set for molding bottles satisfying the target values are deduced.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an operating condition inference device for a blow molding machine, an operating condition inference method and program for a blow molding machine, and a learning device.

Background Art

[0002] Many resin bottles typified by PET bottles are made by first manufacturing a preform having a shape similar to a test tube and then performing stretch blow molding on the preform. A blow molding machine for stretch blow molding is used to stretch the preform and shape the bottle (see Patent Document 1 for the blow molding machine).

[0003] A wide variety of setting items are set for the blow molding machine. In recent years, the diversification of bottle types has advanced, and along with the demands for weight reduction and utilization of recycled materials, the diversification of preforms has also advanced. If the type of bottle or the type of preform for molding the bottle is different, the settings for the blow molding machine naturally need to be adjusted (changed). The setting work of the blow molding machine often relies on the intuition and experience of skilled operators.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object of the present invention is to assist various settings of a blow molding machine.

[0006] The operating condition inference device for a blow molding machine according to the present invention heats a preform with a heater, moves the heated preform to a mold, inserts a stretching rod into the preform to axially stretch the preform, supplies a pressurized fluid inside to radially stretch the preform, and thereby forms a bottle with a hollow inside. The inference device infers the operating conditions of the blow molding machine, and includes a learned model in which parameters are learned so that the physical property information of the preform is input data and the operating conditions of the blow molding machine are output data, and the inference means for inferring the operating conditions of the blow molding machine to be set in the blow molding machine to form a bottle satisfying the target value, which is inferred by inputting the target value of the physical property information that the bottle after molding should satisfy into the learned model.

[0007] According to the present invention, the operating conditions of the blow molding machine to be set in the blow molding machine to form a bottle satisfying the target value are inferred. The inferred operating conditions are based on a learned model in which parameters are learned so that the physical property information of the bottle is input data and the operating conditions of the blow molding machine are output data. Without depending on the operator's experience, the operating conditions of the blow molding machine for forming a bottle having the target physical property information can be inferred in a short time.

[0008] Preferably, the input data further includes at least one of the physical property information of the preform, the molding conditions of the bottle, and the environmental information during the manufacture of the bottle. The possibility of inferring the operating conditions of the blow molding machine for forming a bottle having the target physical property information more accurately and with high precision is enhanced.

[0009] In one embodiment, the inferred operating conditions of the blow molding machine are the heater output conditions for heating the preform and the pressure conditions of the fluid supplied to the preform. This is because the heater output and the fluid pressure (blow pressure) have a great influence on the physical properties of the bottle formed by the blow molding machine. The learned model infers the heater output and the fluid pressure that should be set for the blow molding machine to form a bottle that satisfies the target physical property values. It is possible to increase the certainty of being able to form a bottle that meets a predetermined physical property value target without depending on the operator's experience.

[0010] The inferred operating conditions of the blow molding machine described above may further include, in addition to the heater output and the fluid pressure (blow pressure) described above, the timing conditions for supplying fluid to the preform and / or the speed conditions of the stretching rod inserted into the preform.

[0011] Preferably, the input data includes information regarding the presence or absence of appearance abnormalities of the bottle, and the inference means infers the operating conditions of the blow molding machine that should be set for the blow molding machine to form a bottle that satisfies the target value and has no appearance abnormalities. It is possible to infer the operating conditions for forming a higher-quality bottle.

[0012] In one embodiment, the physical property information of the preform includes data regarding the material and weight of the preform.

[0013] In other embodiments, the molding conditions of the bottle include data regarding the volume and shape of the bottle.

[0014] In other embodiments, the physical property information of the bottle includes at least one of the filled volume, buckling strength, overall height, wall thickness, and section weight of the bottle.

[0015] In other embodiments, the environmental information includes at least one of the temperature and humidity around the blow molding machine.

[0016] The present invention also provides a learning device suitable for generating a learned model for use in a blow operation condition inference device. The learning device according to the present invention heats a preform with a heater, moves the heated preform to a mold, inserts a stretching rod into the preform to stretch the preform in the axial direction, supplies a pressurized fluid inside to stretch the preform in the outward direction, thereby forming a bottle with a hollow inside. A learning data acquisition means for acquiring learning data including the operating conditions of the blow molding machine and the physical property information of the bottle, and using the above learning data, from the target value of the physical property information that the molded bottle should satisfy, the above blow molding machine should be set to form a bottle that satisfies the above target value. It is provided with a model generation means for generating a learned model for inferring the operating conditions of the blow molding machine.

[0017] The present invention further provides a method for inferring the operating conditions of a blow molding machine and a program for causing a computer to execute this method.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Examples

[0019] FIG. 1 is a plan view schematically showing a blow molding machine.

[0020] A large number of resin preforms P, for example, made of PET (polyethylene terephthalate), are prepared at the preform supply station 1. The preforms P are supplied from the preform supply station 1 to the heating station 2 along the conveying path. The heating station 2 is provided with one or a plurality of heater zones 5 (six heater zones 5 are shown in FIG. 1). By passing through the heater zone 5, the preform P is heated to a blow-moldable temperature. The number of the heater zones 5 can be arbitrarily designed.

[0021] The preform P heated at the heating station 2 proceeds to the molding station 3. The molding station 3 includes a rotating wheel 6 and a plurality of molds 7 arranged along the rotating wheel 6 and rotating around the molding station 3 at a constant speed according to the circumferential movement of the rotating wheel 6. The preforms P supplied to the molding station 3 are respectively set in the molds 7. Note that the molding station 3 does not necessarily need to include the rotating wheel 6, and the molds 7 may be conveyed linearly. The number of the molds 7 provided in the molding station 3 is also arbitrary.

[0022] The stretching rod is inserted into the preform P from the mouth of the preform P set in the mold 7, whereby the preform P is stretched in the axial direction (longitudinal direction). Fluid, typically compressed air, is blown into the preform P from the mouth of the axially stretched preform P (air blow). By the air blow, the preform P is stretched outward (axial direction and circumferential (diameter) direction). The air blow is performed in two steps. First, the preform P is inflated by blowing air with a weak pressure (hereinafter referred to as the first air), and then, by blowing air with a high pressure (hereinafter referred to as the second air), the preform P is made to adhere to the inner surface of the mold, whereby the bottle B having a shape conforming to the shape of the mold 7 is produced. After the compressed air and the stretching rod in the bottle B are removed, the bottle B is taken out from the mold 7. The bottle B exiting the molding station 3 proceeds to subsequent processes (not shown), such as a sterilization process, a beverage filling process, a cap attaching process, etc., whereby the beverage-filled bottle is completed.

[0023] Figure 2 schematically shows the positional relationship between the preform P and the heater zone 5 when being conveyed through the heating station 2, and the cross-section of the heater zone 5 is schematically shown. Also shown in Figure 2 is the spindle 8 for holding and rotating the preform P. In the heating station 2, the preform P is conveyed through the heating station 2 while rotating about its axial direction as the rotation axis by the spindle 8.

[0024] The heating station 2 provided in the blow molding machine includes the heater zone 5 as described above, where the preform P is softened so that the preform P can be stretched by the stretching rod and air blow. When a plurality of heater zones 5 are provided, there are types that heat the entire preform P, types that intensively heat the periphery of the support ring of the preform P (focus heater), types that blow air over the entire preform P without a heater for heating (surface screening), and types that blow air onto a part of the preform P (air knife), which are also included in the heater zone 5. Various types can be used, and thus the heating of the preform P can be finely controlled. The heater zone 5 shown in FIG. 2 is of the type that heats the entire preform P. The plurality of heater zones 5 can be controlled independently of each other.

[0025] The heater zone 5 shown in FIG. 2 includes a plurality (seven in FIG. 2) of heaters 5a (for example, halogen lamps) arranged at intervals along the axial direction of the preform P. Each of the seven heaters 5a included in the heater zone 5 can be controlled independently of each other. The number of heaters 5a included in the heater zone 5 can also be arbitrary.

[0026] For example, when it is desired to partially increase the wall thickness of the bottle B formed by the blow molding machine, it is conceivable to lower the heating temperature of the part of the preform P corresponding to that part. For example, using the heater zone 5 of the type shown in FIG. 2, by lowering the output of the heater 5a that heats that part, the heating temperature of a part of the preform P can be lowered, and a bottle B with a partially thick wall can be formed.

[0027] FIG. 3 schematically shows the state of blow molding and shows a longitudinal section of the mold 7 in which the preform P is set.

[0028] The preform P heated at the heating station 2 and transferred to the forming station 3 is surrounded by split molds 7a and 7b that make up the mold 7 on its lateral periphery. The lower surfaces of the split molds 7a and 7b are closed by the bottom mold 7c, and a valve block 13 is joined to the mouth of the preform P. A stretching rod 11 is inserted into the preform P through a hole formed in the valve block 13. The preform P is axially stretched by the stretching rod 11.

[0029] The valve block 13 is provided with two valves V1 and V2 inside it. Compressed air from the air pump 12 is fed into the preform P through the valve V1. The above-mentioned first air and second air are sequentially blown into the preform P from the valve V1, whereby the preform P bulges outward and the bottle B is formed.

[0030] After the bottle B is formed, the compressed air inside the bottle B is exhausted outside through the valve V2 of the valve block 13. The stretching rod 11 is withdrawn from the bottle B, and the bottom mold 7c and the split molds 7a and 7b are opened, whereby the completed bottle B is carried out of the blow molding machine.

[0031] At the forming station 3, the rod speed (advancing speed) of the stretching rod 11, the timing of starting to blow the first air, the pressure of the first air, the timing of switching from the first air to the second air, the pressure of the second air, etc. can be controlled.

[0032] For example, when it is desired to reduce the wall thickness near the bottom of the bottle B formed by the blow molding machine, it is conceivable to advance the timing of starting to blow the first air. Since the first air is supplied before the stretching rod is fully extended (before the preform P is fully stretched vertically), the wall thickness at the bottom of the bottle B becomes thinner, and instead, the wall thickness from the neck to the shoulder of the bottle B can be increased.

[0033] Figure 4 shows a functional block diagram of the blow molding machine.

[0034] The blow molding machine is functionally roughly divided into a heater function 20 and a blow function 30. The heater function 20 is realized by a plurality of heater zones 5 provided in the heating station 2 described above. The blow function 30 is realized by a stretching rod 11, an air pump 12, a valve block 13, etc. provided in the molding station 3.

[0035] As described above, each of the plurality of heater zones 5 provided in the heating station 2, and further each of the plurality of heaters 5a provided in the heater zone 5, can be controlled individually. In the molding station 3, the pressure of the air supplied from the air pump 12, the supply timing of the air, and the rod speed (entry speed) of the stretching rod 11 can also be controlled. The heater function 20 and the blow function 30 are centrally managed by the control device 40.

[0036] A setting device 50 is connected to the control device 40. The setting device 50 is provided with a display device (not shown in the figure), and a number of setting items are displayed on the setting screen displayed on the display device. According to the setting of a number of setting items using the setting screen, the control device 40 controls the heater function 20 and the blow function 30.

[0037] The setting device 50 can set the setting values of a number of setting items, and according to these a number of setting values, the heater function 20 and the blow function 30 of the blow molding machine are finely controlled. By changing the setting value, as described above, for example, the wall thickness of a specific part of the finally molded bottle B can be made thicker or thinner.

[0038] Specifically, the setting device 50 is set with preform property information such as the material and weight of the preform P, bottle forming conditions such as the capacity and shape of the bottle B to be formed, heater setting information such as the output of the heater 5a in the heater zone 5, the rod speed (advancing (inserting) speed) of the stretching rod 11, the timing to start the first air, the pressure of the first air, the timing to switch from the first air to the second air, blow setting information such as the pressure of the second air, and the like. There are many setting items for the blow molding machine, and at the site, a skilled operator finely adjusts the setting values of a large number of setting items. This is because the skilled operator has experience in how to adjust any of the many setting items and how the result will appear on the bottle B.

[0039] In order to mechanically realize the fine adjustment of the setting items by a skilled operator and reduce the adjustment work time and stabilize the quality of the formed bottle B, in this embodiment, machine learning is performed as described below. A target value of the property information that the formed bottle B should satisfy is given, and the operating conditions set for the blow molding machine to form the bottle B that satisfies the target value are inferred as a result of the machine learning process.

[0040] As described below, the operating conditions inferred using the machine learning process are the heater output and the blow pressure in the blow molding machine in this embodiment. In addition to these, the blow timing and the rod speed may also be added to the inferred operating conditions. This is because the heater output, the blow pressure, the blow timing, and the rod speed have a correlation with the physical properties of the bottle B formed from the preform P by the blow molding machine.

[0041] FIG. 5 shows a block diagram of the learning device 70. The learning device 70 performs machine learning related to a blow molding machine, and includes a learning data acquisition unit 71 that acquires learning data which is data used for learning, a model generation unit 72 that generates a learning model used to infer the heater output, blow output, blow timing, and rod speed of the blow molding machine, and a learned model 73 (its storage unit) that is generated. The learning device 70 is realized by, for example, a computer device and operates according to a program that causes the computer device to realize the functions of the learning device 70.

[0042] The learning data acquisition unit 71 acquires (receives) as learning data data that was set or measured when actually forming the bottle B through a network (for example, the Internet) from a plurality of bottle molding factories (beverage manufacturing factories with bottles) where the blow molding machine is installed. The learning data includes the following data.

[0043] (1) Heater output It is the output of a plurality of heater zones 5 provided in the heating station 2 of the blow molding machine. The data representing the heater output may be the output ratio (total output) with respect to the maximum output of the entire heating station 2, each heater zone 5, or each heater 5a (for example, 80% etc.), or may be the on / off data of a plurality of heaters 5a provided in the heater zone 5. (2) Blow pressure It is the pressure of the first air and the pressure value of the second air blown into the preform P.

[0044] (3) Blow timing It is the start timing of the first air and the timing (switching timing) when switching from the first air to the second air. Instead of the switching timing, a combination of the end timing of the first air and the start timing of the second air may be used. As the data representing the blow timing, the rotation angle of the rotary wheel 6 with the position where the preform P is delivered to the forming station 3 as a reference (0°) can be used. For example, if the start timing of the first air is "33°", it means that the first air is started at the timing when the preform P is delivered to the forming station 3 and rotated by 33° by the rotary wheel 6.

[0045] (4) Rod speed It is the traveling (insertion) speed of the stretching rod 11 inserted into the preform P.

[0046] (5) Preform property information It is the material, weight, dimensions, etc. of the preform P. The dimensions generally include the diameter of the opening of the preform P, the diameter of the body, and the length from the support ring to the lower end (body length).

[0047] (6) Conditions during bottle forming It is the capacity and shape of the bottle B. The shape is generally distinguished into round and square.

[0048] (7) Environmental information It is the measured values of the temperature and humidity around the blow molding machine.

[0049] (8) Bottle property information It is the measured values of the filling capacity, buckling strength, overall height, wall thickness, and section weight (the weight of each part divided by laterally cutting (cutting perpendicular to the axial direction) the bottle B at one or a plurality of predetermined positions) of the formed bottle B. For the wall thickness and section weight, the respective values at a plurality of positions of the bottle B, such as the neck, shoulder, body, and bottom (even finer ones may be used) are generally used.

[0050] (9) Bottle appearance information The bottle B formed from the preform P may become white and turbid (whitened). The presence or absence of this whitening is included in the bottle appearance information. The presence or absence of structural defects in the bottle B may also be included in the bottle appearance information. There are two types of whitening: whitening due to overstretching and whitening due to crystallization. Whitening due to overstretching is likely to occur when the heating of the preform P is insufficient, and whitening due to crystallization occurs when the preform P is overheated and the material of the preform P crystallizes.

[0051] The model generation unit 72 learns the learning data from the learning data acquisition unit 71 and generates a learned model for inferring the operating conditions to be set in the blow molding machine in order to mold the bottle B that meets the target, specifically, the heater output, blow pressure, blow timing, and rod speed.

[0052] As the learning algorithm used by the model generation unit 72, known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used.

[0053] For example, LightGBM, which is a learning algorithm with high prediction accuracy and short training time, can be used. The learned model 73 represents the relationship between the explanatory variables (specifically, the bottle physical properties information, preform physical properties information, bottle molding conditions, and environmental information) and the target variables (the heater output, blow pressure, blow timing, and rod speed of the blow molding machine), and is obtained by learning the parameters for calculating the values of the target variables. External setting variables (hyperparameters) for managing the training of the learning model (such as the number of branches of the decision tree) are optimized by, for example, a Bayesian optimization library.

[0054] The output data (objective variable) of the learned model 73 may preferably be selected preferentially those that have a great influence on the physical properties of the bottle B formed by the blow molding machine. Here, an example in which four data, namely, the heater output, the blow pressure, the blow timing, and the rod speed of the blow molding machine, are treated as the output data will be particularly described. For example, two data, namely, the heater output and the blow pressure (particularly the pressure of the first air), may be used as the output data (objective variable). In LightGBM, since the importance (degree of influence) of the explanatory variables that affect the objective variable is calculated for each explanatory variable, when paying attention to a bottle physical property of interest (for example, the full filling volume) (one of the explanatory variables), data (objective variable) that has a great influence on the bottle physical property of interest may be selected as the output data of the learned model 73.

[0055] Regarding the input data (explanatory variables) of the learned model 73 as well, it is of course possible to limit them to, for example, the physical property information (for example, the full filling volume, which is one of them) of the bottle B formed by the blow molding machine. However, when using a plurality of types of input data, the possibility of inferring setting items and setting values that the operator does not notice increases.

[0056] FIG. 6 is a block diagram of an inference device for inferring the operating conditions to be set in the blow molding machine.

[0057] The inference device 80 includes an inference data acquisition unit 82 and an inference unit 83. The inference device 80 is also realized by, for example, a computer device and operates according to a program that causes the computer device to realize the functions of the inference device 80.

[0058] The inference data acquisition unit 82 is given at least the target value (design value) (at least one of the full filling volume, buckling strength, overall height, wall thickness, and section weight) 81a of the physical property information of the bottle B to be formed, and preferably the physical property information of the preform P used for forming the bottle B (at least one of the material and weight of the preform P), the bottle forming conditions (at least one of the bottle volume and shape), and the environmental information (at least one of the temperature and humidity) 81b.

[0059] The inference unit 83 uses the learned model 73 to infer the operating conditions of the blow molding machine, here the heater output, blow pressure, blow timing, and rod speed. That is, the inference unit 83 inputs the inference data acquired by the inference data acquisition unit 82 into the learned model 73, and outputs the heater output, blow pressure, blow timing, and rod speed such that the generated bottle will have the target physical property information.

[0060] The target (value) of the physical property information of the bottle to be formed may be a specific numerical value or a numerical range with a predetermined width. For the wall thickness and section weight, the values at a plurality of locations on the bottle B, for example, the neck, shoulder, body, and bottom, are generally set.

[0061] By using the learned model 73 that has been trained using a large number of learning data, the inference device 80 can output the heater output, blow pressure, blow timing, and rod speed to be set in the setting device 50 of the blow molding machine in order to generate a bottle B having physical properties that meet the target or fall within the target range. By setting the heater output, blow pressure, blow timing, and rod speed output from the inference device 80 in the setting device 50 of the blow molding machine, it is possible to manufacture a bottle B that meets the target physical properties as if set by a skilled operator.

[0062] As described above, since the learning data includes bottle appearance information, the inference device 80 can also output the heater output, blow pressure, blow timing, and rod speed to be set in the setting device 50 of the blow molding machine in order to generate a bottle B having the target physical properties and no appearance abnormalities (for example, no whitening). Higher quality bottles B can be formed.

[0063] FIG. 7 shows the processing result of an inference device 80 using a learned model 73, where "bottle capacity (600 ml)" and "bottle shape (round shape)" are used as the conditions during bottle molding, the "filling capacity" (ml) is set as the target value (explanatory variable, input data) of the physical property information of the bottle to be molded, and the operating conditions of two blow molding machines, namely "heater output" (%) (total output) and "blow pressure" (pressure of the first air) (bar), are set as the inferred values (objective variable, output data). In FIG. 7, when the target values of the filling capacity are varied, the heater output and blow pressure inferred are set for the blow molding machine, and the measured values of the filling capacity of each of the five bottles B actually molded from five preforms P of the same type, as well as the error between the target value and the measured value, are also shown. In the "error between the target value and the measured value", "error (= measured value - target value)", "error rate (= error ÷ measured value)", "absolute error rate (= absolute value of the error rate)", and "average absolute error rate (= total of the absolute error rates ÷ number of data)" are also shown. FIG. 8 is a graph with the target value of the filling capacity on the horizontal axis and the measured value of the filling capacity on the vertical axis, and the target values and measured values of each of the five data are plotted by circles. The broken line shown in the graph of FIG. 8 is an auxiliary line indicating the case where the target value and the measured value match.

[0064] The average absolute error rate of the five data is as small as 0.124%. By forming the bottle B using the heater output and blow pressure inferred using the learned model 73, it was confirmed that a bottle B with a filling capacity very close to the target value can be formed.

Explanation of Signs

[0065] 1 Preform supply station 2 Heating station 3 Molding station 5 Heater zone 5a Heater 7 Mold 11 Stretch rod 12 Air pump 13 Valve block 20 Heater function 30 Blow function 40 Control device 50 Setting device 70 Learning device 71 Learning data acquisition unit 72 Model generation unit 73 Learned model 80 Inference device 82 Inference data acquisition unit 83 Inference unit B Bottle P Preform

Claims

1. 1. An inference device for inferring operating conditions of a blow molding machine which heats a preform with a heater, moves the heated preform to a mold, inserts a stretch rod into the preform to stretch the preform in an axial direction, and supplies a pressurized fluid to the inside of the preform to stretch the preform outward, thereby forming a hollow bottle, A trained model in which parameters are trained to use the physical property information of a bottle as input data and the operating conditions of a blow molding machine as output data; and The apparatus further includes an inference means for inferring operating conditions of the blow molding machine that should be set on the blow molding machine in order to mold a bottle that satisfies a target value, the operating conditions being inferred by inputting a target value of physical property information that the molded bottle should satisfy into the trained model, An operating condition inference device for blow molding machines.

2. The input data further includes at least one of physical property information of the preform, molding conditions of the bottle, and environmental information during the manufacture of the bottle. The operating condition inference device for a blow molding machine according to claim 1 .

3. The operating conditions of the blow molding machine are a heater output condition for heating the preform and a pressure condition of a fluid supplied to the preform. The operating condition inference device for a blow molding machine according to claim 1 .

4. The operating conditions of the blow molding machine further include timing conditions for supplying fluid to the preform; The operating condition inference device for a blow molding machine according to claim 3.

5. The operating conditions of the blow molding machine further include a speed condition of a stretch rod inserted into the preform; The operating condition inference device for a blow molding machine according to claim 3 or 4.

6. The input data includes information regarding the presence or absence of an appearance abnormality of the bottle, The inference means is Inferring the operating conditions of the blow molding machine that should be set for the blow molding machine in order to mold a bottle that satisfies the target value and has no abnormality in appearance. The operating condition inference device for a blow molding machine according to claim 1 .

7. The physical property information of the preform includes data on the material and weight of the preform, The operating condition inference device for a blow molding machine according to claim 2.

8. The above bottle molding conditions include data on the capacity and shape of the bottle. The operating condition inference device for a blow molding machine according to claim 2.

9. The physical property information of the bottle includes at least one of the full capacity, buckling strength, total height, wall thickness, and section weight of the bottle. The operating condition inference device for a blow molding machine according to claim 1 .

10. The environmental information includes at least one of the temperature and humidity around the blow molding machine. The operating condition inference device for a blow molding machine according to claim 2.

11. a learning data acquisition means for acquiring learning data including operating conditions of a blow molding machine that heats a preform with a heater, moves the heated preform to a mold, inserts a stretching rod into the preform to stretch the preform in the axial direction, and supplies a pressurized fluid to the inside to stretch the preform outwardly, thereby forming a hollow bottle, and physical property information of the molded bottle; and a model generation means for generating a trained model for inferring, from target values ​​of physical property information that a molded bottle should satisfy, operating conditions of the blow molding machine that should be set in the blow molding machine in order to mold a bottle that satisfies the target values, using the training data; Learning device.

12. 1. A method for predicting operating conditions of a blow molding machine which heats a preform with a heater, moves the heated preform to a mold, inserts a stretch rod into the preform to stretch the preform in an axial direction, and supplies a pressurized fluid therein to stretch the preform outwardly, thereby forming a hollow bottle, comprising: A trained model is prepared in which parameters are trained so that the physical property information of the bottle is used as input data and the operating conditions of the blow molding machine are used as output data. Inferring operating conditions of the blow molding machine that should be set on the blow molding machine in order to mold a bottle that satisfies the target value, which is inferred by inputting target values ​​of physical property information that the molded bottle should satisfy into the trained model; A method for inferring operating conditions of a blow molding machine.

13. A program for causing a computer to execute the method according to claim 12.

Images