Prediction system, program, and prediction method

Abstract

SOLUTION: The present invention provides a prediction system for predicting physical properties of a foamed resin molded article molded in a mold, the prediction system including a molding prediction part that predicts the physical properties of the foamed resin molded article molded in the mold by inputting at least one of mold conditions related to a connection between inside and outside of the mold used for molding and molding conditions related to at least the conditions applied in a molding process. The present invention also provides a program that is executed by a computer and causes the computer to function as the prediction system. The present invention also provides a prediction method using the prediction system.SELECTED DRAWING: Figure 1D

Description

[Technical Field]

[0001] The present invention relates to a prediction system, a program, and a prediction method. [Background technology]

[0002] Patent Document 1 describes a method for determining design parameters for an injection-molded product, comprising: (1) a design parameter setting step for selecting design parameters that optimize the injection-molded product; (2) a design condition setting step for setting design conditions for the design parameters; (3) an objective function setting step for setting an analytical value obtained by an injection molding simulation as an objective function; (4) an objective condition setting step for setting objective conditions under which the objective function related to the pressure is appropriate; (5) a calculation model creation step for creating a calculation model in which the shapes of the molded product and the runner are divided into a plurality of infinitesimal elements; (6) an injection molding simulation step for calculating an analytical value of the objective function using the design parameters and the calculation model; (7) a determination step for determining whether the analytical value of the objective function satisfies the objective conditions; (8) a design parameter control step for changing the design parameters within the range of the design conditions when the analytical value of the objective function does not satisfy the objective conditions; and (9) a design parameter control step for determining whether the analytical value of the objective function satisfies the objective conditions at an initially set number of gate points. (10) a gate adding step of adding a new gate near the center of a low-pressure region in the pressure distribution inside the mold during molding, obtained by the injection molding simulation using the selected provisional optimal conditions; (11) a design parameter resetting step of adding the gate added in the gate adding step to the design parameters; and (12) a design parameter resetting step of adopting the low-pressure region in the pressure distribution inside the mold during molding, obtained by the injection molding simulation using the selected provisional optimal conditions, as a position search range for the gate added in the gate adding step, and resetting the design conditions, wherein at least each of the steps (6) to (12) is repeated until it is determined in the judging step that the analysis value satisfies the target condition (Claim 1). Patent Document 2 describes that the method includes "a mold design method characterized by determining in advance mold design parameters related to the arrangement, shape, and / or dimensions of the resin inlet channels by combining a numerical analysis method for calculating the injection molding process with a computer-aided optimization method, in order to obtain suitable injection molding conditions when injection molding is performed using a mold having multiple resin inlet channels into the cavity" (Claim 1). [Prior art document] [Patent Documents] [Patent Document 1] Japanese Patent Application Laid-Open No. 2003-080575 [Patent Document 2] Japanese Patent Application Laid-Open No. 2005-007859 Summary of the Invention

[0003] In a first aspect of the present invention, there is provided a prediction system for predicting the physical properties of a foamed resin molded article molded in a mold. The prediction system includes a molding prediction unit. The molding prediction unit inputs at least one of mold conditions related to the connection between the inside and outside of the mold used for molding and molding conditions related to at least the conditions applied in the molding process, and predicts the physical properties of the foamed resin molded article molded in the mold.

[0004] In the above, the molding prediction unit may input both the mold conditions and the molding conditions to predict the physical properties of the foamed resin molded article.

[0005] In the above, the mold conditions may include at least one of ventilation conditions, such as the number, position, shape, and size of ventilation holes and core vents that connect the inside and outside of the mold and allow gas to pass through, and feeder attachment port conditions, such as the number, position, shape, and size of feeder attachment ports that supply raw material for the foamed resin molded body to the mold.

[0006] In the above, the molding conditions may include at least one of information on heating temperature, heating time, vapor pressure, cooling temperature, cooling time, raw material filling time, filling air pressure, and cracking amount.

[0007] In the above, the molding prediction unit may further input raw material information on the raw materials of the foamed resin molded body to predict the physical properties of the foamed resin molded body.

[0008] In the above, the molding prediction unit may further input design information relating to the design shape of the foamed resin molded body to predict the physical properties of the foamed resin molded body.

[0009] In the above, the molding prediction unit may predict one or more of the density, compressive strength, tensile strength, flexural strength, water absorption, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body as physical properties of the foamed resin molded body.

[0010] In the above, the molding prediction unit may set a plurality of regions by dividing the foamed resin molded body, and predict the physical properties of the foamed resin molded body for each of the regions.

[0011] The above-mentioned method may further comprise a property evaluation unit that determines whether or not the predicted property of the foamed resin molded article exceeds a predetermined standard.

[0012] In the above, a search unit may be further provided that uses the prediction result of the forming prediction unit to search for mold conditions and / or forming conditions that satisfy a predetermined standard.

[0013] In the above, the prediction system may further include a learning unit. The learning unit may use learning data to re-learn the molding prediction model used by the molding prediction unit for prediction. The learning data may include proposed conditions, which are mold conditions and / or molding conditions proposed by the search unit, and physical properties associated with a foamed resin molded body actually molded using the proposed conditions. The molding prediction unit may predict physical properties using the molding prediction model re-learned by the learning unit.

[0014] In the above, the prediction system may further include a surface prediction unit. The surface prediction unit may predict the physical properties of each region of the foamed resin molded body based on photographed images of the surface of the foamed resin molded body. The physical properties included in the training data may be physical properties predicted by the surface prediction unit from photographed images of actually molded foamed resin molded bodies.

[0015] In a second aspect of the present invention, there is provided a prediction system including a surface prediction unit, which may predict physical properties of each region of a foamed resin molded body based on a captured image of the surface of the foamed resin molded body.

[0016] In the above, the surface prediction unit may predict one or more of the physical properties of the foamed resin molded body, including the density, compressive strength, tensile strength, flexural strength, water absorption, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body.

[0017] In the above, the surface prediction unit may predict the physical properties based on an image captured using reflected light and / or transmitted light of the foamed resin molded article.

[0018] In the above, the surface prediction section may predict the physical properties based on a luminance image of the foamed resin molded body.

[0019] In the above, the surface prediction unit may predict the physical properties using a binarized image obtained by binarizing the luminance image.

[0020] In a third aspect of the present invention, there is provided a prediction method using the above prediction system. The prediction method may include a molding prediction step. In the molding prediction step, at least one of mold conditions and molding conditions may be input to predict the physical properties of a foamed resin molded body.

[0021] In a fourth aspect, the present invention provides a program that, when executed by a computer, causes the computer to function as the above-described prediction system.

[0022] The above summary of the invention does not list all of the features of the present invention, and subcombinations of these features may also constitute inventions. [Brief explanation of the drawings]

[0023] [Figure 1A] An example of a mold foam molding process is shown below. [Figure 1B] An example of a mold foam molding process is shown below. [Figure 1C] An example of a mold foam molding process is shown below. [Figure 1D] An example of a mold foam molding process is shown below. [Figure 2] 1 shows the configuration of a prediction system 100 according to this embodiment. [Figure 3] 1 shows a flow of prediction by the prediction system 100 according to this embodiment. [Figure 4] An example of input and output of information by the forming prediction model 50 is shown below. [Figure 5] An example of input and output of information by the surface prediction model 60 is shown below. [Figure 6] 10 shows an example of a sub-flow of S200 according to this embodiment. [Figure 7] An example of processing by the forming prediction unit 120 in the subflow of S200 in FIG. 6 will be described below. [Figure 8A] An example of the feature vector generated in S240 is shown below. [Figure 8B] An example of the physical properties predicted in S250 is shown below. [Figure 9] 10 shows an example of a subflow of the surface prediction unit 130 acquiring physical properties. [Figure 10] An example of processing by the surface prediction unit 130 in the subflow of S400 in FIG. 9 will be described below. [Figure 11] An example of the captured image processed in S415 is shown below. [Figure 12] 22 illustrates an example computer 2200 in which aspects of the present invention may be embodied, in whole or in part. DETAILED DESCRIPTION OF THE INVENTION

[0024] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0025] 1A to 1D show an example of a mold foam molding process. Foamed resin molded products can be molded into various shapes and are known to be lightweight and have excellent heat insulating properties. Foamed resin molded products can be produced by foaming and molding resin beads, which are the raw material.

[0026] 1A, foamed resin beads 12, which are raw material, are first injected into the region formed between mold 22 and mold 24 via feeder 14 (raw material supply port) attached to mold 22. Mold 22 and mold 24 may not be completely closed, but may be open by a cracking amount C.

[0027] The expanded resin beads may be, for example, expanded beads obtained by expanding a resin such as modified polyphenylene ether using a high-pressure gas such as carbon dioxide gas. As an example, the expanded resin beads 12 may be SunForce (registered trademark) manufactured by Asahi Kasei.

[0028] Next, the mold 22 and the mold 24 are completely closed as shown in Figure 1B, thereby forming a cavity 30 of a predetermined shape.

[0029] 1C, the inside of the space 30 is heated by a high-temperature gas such as steam, which causes secondary expansion of the foamed resin beads 12, filling the gaps between the beads and forming a foamed resin molded body 32 having approximately the same shape as the space 30.

[0030] 1D, the mold 22 and the mold 24 are separated after cooling, etc., and the foamed resin molded body 32 is thereby removed from the mold 24.

[0031] In this way, the foamed resin molded body 32 can be molded into a relatively free shape using a mold. However, depending on the shape of the molded body, the quality of the molded body may not be maintained constant. For example, depending on the shape of the mold, raw materials may not be supplied appropriately. For example, if the molded body includes a thin portion (e.g., a portion less than 3 mm thick), the foamed resin beads may not be uniformly dispersed in that portion of the mold, which may affect the quality of the molded body.

[0032] FIG. 2 shows the configuration of a prediction system 100 according to this embodiment. The prediction system 100 predicts physical properties, such as density or strength, of a foamed resin molded body based on manufacturing conditions, such as mold conditions and molding conditions, used for molding. The prediction system 100 proposes conditions under which desired physical properties can be obtained. Furthermore, the prediction system 100 predicts the physical properties of each region of the foamed resin molded body based on captured images of the surface of the foamed resin molded body. The prediction system 100 includes a learning unit 110, a molding prediction unit 120, a surface prediction unit 130, a search unit 140, a physical property evaluation unit 150, and a proposal unit 170.

[0033] The learning unit 110 learns a prediction model used for prediction by the prediction system 100. For example, the learning unit 110 learns a forming prediction model and / or a surface prediction model.

[0034] The molding prediction model predicts the physical properties of a foamed resin molded body molded in a mold by inputting conditions and / or information related to mold molding. For example, the molding prediction model predicts the physical properties of a foamed resin molded body molded in a mold by inputting at least one of mold conditions related to the connection between the inside and outside of the mold used for molding and molding conditions related to at least the conditions applied in the molding process.

[0035] The surface prediction model predicts the physical properties of each region of a foamed resin molded body based on the appearance of the foamed resin molded body. For example, the surface prediction model predicts the physical properties of each region of a foamed resin molded body based on a photographed image of the surface of the foamed resin molded body.

[0036] The molding prediction unit 120 makes predictions using the molding prediction model learned by the learning unit 110. For example, the molding prediction unit 120 inputs at least one of mold conditions related to the connection between the inside and outside of the mold used for molding and molding conditions related to at least the conditions applied in the molding process, and predicts the physical properties of the foamed resin molded body molded in the mold. For example, the molding prediction unit 120 inputs both the mold conditions and the molding conditions and predicts the physical properties of the foamed resin molded body.

[0037] The molding prediction unit 120 may predict the physical properties of a foamed resin molded body by inputting, in addition to the mold conditions and molding conditions, raw material information on the raw materials of the foamed resin molded body into the molding prediction model. The molding prediction unit 120 may predict the physical properties of a foamed resin molded body by inputting, in addition to the mold conditions and molding conditions, design information on the design shape of the foamed resin molded body into the molding prediction model.

[0038] The molding prediction unit 120 may set regions by dividing the foamed resin molded body into a plurality of parts, and predict the physical properties of the foamed resin molded body for each of the regions. Additionally / alternatively, the molding prediction unit 120 may predict the overall physical properties of the foamed resin molded body.

[0039] The surface prediction unit 130 makes predictions using the surface prediction model learned by the learning unit 110. The surface prediction unit 130 predicts the physical properties of the foamed resin molded body based on a captured image of the surface of the foamed resin molded body. For example, the surface prediction unit 130 may predict the physical properties of the foamed resin molded body as a whole or for each region.

[0040] The search unit 140 searches for mold conditions and / or molding conditions that satisfy predetermined criteria using the prediction results of the molding prediction unit 120. For example, the search unit 140 generates search conditions such as mold conditions and / or molding conditions to be used in the search.

[0041] The physical property evaluation unit 150 determines whether the predicted physical properties of the foamed resin molded body exceed a predetermined standard, and then determines whether or not the search by the search unit 140 should continue.

[0042] The proposing unit 170 outputs the mold conditions and / or molding conditions for the foamed resin molded product that are determined by the property evaluating unit 150 to satisfy the predetermined criteria as proposed mold conditions and / or proposed molding conditions.

[0043] The prediction system 100 may be a computer such as a PC (personal computer), a tablet computer, a smartphone, a workstation, a server computer, or a general-purpose computer, or may be a computer system in which a plurality of computers are connected.

[0044] Alternatively, the prediction system 100 may be a dedicated computer designed for each application, or may be dedicated hardware implemented using dedicated circuits. The prediction system 100 may be implemented by a single device (computer), or may be implemented by multiple devices with different roles. For example, the prediction system 100 may be a server-client system that provides cloud services to users who use the prediction method. The prediction system 100 may function as a computer by executing a program.

[0045] Although not specifically described below, the prediction system 100 is equipped with a memory / hard disk, etc., in which information necessary for processing is appropriately stored, and the prediction system 100 transmits information between each processing module, such as the learning unit 110, the forming prediction unit 120, the surface prediction unit 130, the search unit 140, the physical property evaluation unit 150, and the proposal unit 170, as necessary.

[0046] As described above, the prediction system 100 of this embodiment can predict the physical properties of a foamed resin molded body based on the mold conditions, molding conditions, etc. Furthermore, the prediction system 100 can propose proposed mold conditions and / or proposed molding conditions that are expected to result in the physical properties of the foamed resin molded body exceeding the standards. This makes it possible to stabilize the quality of the foamed resin molded body.

[0047] 3 shows a flow of prediction by the prediction system 100 according to this embodiment. The prediction system 100 learns a prediction model by executing each process of S100 to S600, and searches for mold conditions and molding conditions using the prediction model. Some of S100 to S600 and / or some of the subflows may be omitted. Other operations may be performed in addition to S100 to S600.

[0048] First, in S100, the learning unit 110 learns a prediction model. The learning unit 110 may learn a plurality of prediction models for each prediction target. For example, the prediction models may include a molding prediction model that predicts the physical properties of a foamed resin molded body to be molded from manufacturing conditions (e.g., mold conditions and / or molding conditions), and a surface prediction model that predicts the physical properties of a foamed resin molded body from a photographed image of the surface of the foamed resin molded body.

[0049] 4 shows an example of information input and output by the molding prediction model 50. The molding prediction model 50 receives molding conditions 52, mold conditions 54, raw material information 56, and design information 58 as input, and predicts physical properties 59 of a foamed resin molded body molded under these conditions.

[0050] [Molding conditions] The molding conditions 52 define conditions related to the molding process of the foamed resin molded body. For example, the molding conditions 52 may include at least one of information on raw material filling time, filling air pressure, heating temperature, heating time, vapor pressure, cooling temperature, cooling time, and cracking amount.

[0051] The raw material filling time and filling air pressure may be the time required to fill raw material into the mold and the air pressure applied during filling in the raw material filling process into the mold described in Fig. 1A. The cracking amount may be the degree of closure between the molds (e.g., C in Fig. 1A) during the raw material filling process into the mold described in Fig. 1A. For example, the cracking amount may be the distance between predetermined positions of the mold.

[0052] The heating temperature and heating time may be the heating temperature and heating time in the heating process of the raw material in the mold described in Fig. 1C. The heating temperature may be the set temperature of the molding device or the measured temperature of the injected high-temperature gas (e.g., steam). The steam pressure may be the pressure of the high-temperature gas used in the heating process.

[0053] The cooling temperature and cooling time may be the cooling temperature and cooling time in the cooling process after heating described in Fig. 1D. The cooling temperature may be the set temperature of the molding device or the measured temperature of the injected coolant (e.g., water).

[0054] The molding conditions 52 are not limited to those mentioned above, and may include other conditions that can be adjusted by the molding device and / or the operator during molding.

[0055] [Mold conditions] Mold conditions 54 specify conditions related to the mold used in molding the foamed resin molded article. For example, mold conditions 54 may include at least one of ventilation conditions, such as the number, position, shape, and size of vent holes and core vents that connect the inside and outside of the mold and allow gas to pass through, and feeder attachment conditions, such as the number, position, shape, and size of feeder attachment ports that supply raw material for the foamed resin molded article to the mold.

[0056] The vent hole serves to supply air, heated gas, steam, moisture, etc. to the mold during heating and cooling, etc. The core vent serves to release air, heated gas, steam, moisture, etc. from the mold during raw material supply, heating and cooling, etc. The vent hole and core vent may be provided separately, or the same type of hole may be provided to perform both functions. The feeder attachment port is used when supplying raw material.

[0057] The positions of the vent holes, core vents, and feeder attachment ports may be expressed by information that at least partially represents the positions of these holes. For example, the positions of the vent holes, core vents, and feeder attachment ports may be expressed by a position on a three-dimensional coordinate system, a position on a two-dimensional coordinate system on a predetermined plane of the mold, a distance and / or orientation from a predetermined position on the mold, a designation of a position in a space or plane divided into lattices, or other notation.

[0058] The shapes of the vent hole, core vent, and feeder attachment port may be information that at least partially represents the shapes of these holes. For example, the shapes of the vent hole, core vent, and feeder attachment port may be three-dimensional information, two-dimensional information in a predetermined cross section, and / or a classification of the hole shape (e.g., classification of perfect circle, ellipse, polygon, and / or rectangle, etc.).

[0059] The sizes of the vent hole, core vent, and feeder attachment port may be information that at least partially represents the size of these holes. For example, the sizes of the vent hole, core vent, and feeder attachment port may be the cross-sectional size (e.g., long side length, short side length, radius, major axis and / or minor axis, etc.), the depth of the hole, etc.

[0060] The mold conditions 54 are not limited to the above, and may include any other conditions related to the mold (for example, the material of the mold or the heat conductivity of the mold, etc.).

[0061] [Raw material information] The raw material information 56 may include information about the raw material itself. For example, the raw material information 56 may include information about the materials contained in the raw material, the composition of the raw material, and / or the physical properties of the raw material. The materials contained in the raw material may include information about all or part of the materials, such as the type of resin contained in the raw material and other additives. The composition of the raw material includes information about the content of each material (e.g., the weight parts of each material) and / or the content ratio. The physical properties of the raw material may include numerical values ​​such as the size of the raw material (e.g., average particle size), the density of the raw material, the molecular weight of the raw material, the molecular weight distribution, and / or hardness.

[0062] [Design information] The design information 58 may include information about the design shape of the foamed resin molded body. For example, the design information 58 may include design drawing data (e.g., CAD data) of the finished foamed resin molded body.

[0063] [Physical Properties] The physical properties 59 are the physical properties of the foamed resin molded body molded under the mold conditions 54, etc. For example, the physical properties 59 may include one or more of the density, compressive strength, tensile strength, flexural strength, water absorption rate, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body.

[0064] The physical property 59 may be a physical property of the foamed resin molded body as a whole. In addition to or instead of this, the physical property 59 may be set for each region of the foamed resin molded body. In this case, a plurality of physical properties 59 are set for the foamed resin molded body.

[0065] The learning unit 110 uses learning data including a plurality of pairs of explanatory variable data and objective variable data, which are physical properties 59, and learning data including at least a portion of the molding conditions 52, mold conditions 54, raw material information 56, and design information 58, or data obtained by processing these, to learn a molding prediction model 50 that predicts objective variable data from the explanatory variable data.

[0066] The molding prediction model 50 may not use some of the molding conditions 52, the mold conditions 54, the raw material information 56, and the design information 58. For example, the molding prediction model 50 may input only the molding conditions 52, the mold conditions 54, and the design information 58. In this case, the raw material information 56 may be treated as a fixed condition incorporated into the prediction model.

[0067] Instead of directly inputting the mold conditions 54 and the design information 58, the molding prediction model 50 may input, as explanatory variable data, a feature vector obtained by processing the mold conditions 54 and the design information 58 as described below. Alternatively, the molding prediction model 50 may directly input the mold conditions 54 and the design information 58. The learning unit 110 may perform machine learning using a known method (for example, a regression model or a deep learning model).

[0068] 5 shows an example of information input and output by the surface prediction model 60. The surface prediction model 60 receives a photographed image 62 of a foamed resin molded body and predicts physical properties 64 for each region of the photographed foamed resin molded body.

[0069] [Photo] The captured image 62 may be an image of the surface of the foamed resin molded body. For example, the captured image 62 may be an image captured using reflected light and / or transmitted light from the foamed resin molded body. For example, the captured image 62 may be a luminance image of the foamed resin molded body, or a binarized image obtained by binarizing a luminance image.

[0070] The captured image may be a large number of images obtained by dividing an image of the surface of the foamed resin molded body into a large number of regions. The captured image may be divided so that adjacent regions do not overlap. Alternatively, the captured image may be divided so that adjacent regions partially overlap.

[0071] [Physical Properties] The physical property 64 is a physical property of the foamed resin molded body that is the subject of the captured image 62. The physical property 64 may be the same as the physical property 59. For example, the physical property 64 may be one or more of the density, compressive strength, tensile strength, flexural strength, water absorption, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body.

[0072] The physical properties 64 may be the physical properties of each region of the foamed resin molded body. For example, similar to a photographed image, the foamed resin molded body is divided into many regions, and the physical properties 64 are output for each divided region.

[0073] The learning unit 110 uses learning data including a plurality of pairs of explanatory variable data obtained by processing at least a part or the entire captured image 62 and objective variable data, which are physical properties 64, to learn a surface prediction model 60 that predicts objective variable data from the explanatory variable data. The surface prediction model 60 may input the captured image 62 divided into a large number of regions and output physical properties 64 corresponding to each of the large number of regions.

[0074] The learning unit 110 may perform machine learning using a known method (for example, a regression model or a deep learning model).

[0075] In S200, the prediction system 100 searches for search conditions such as mold conditions and molding conditions under which physical properties above a certain standard can be obtained when molding is performed.

[0076] 6 shows an example of a sub-flow of S200 according to this embodiment. The prediction system 100 may search for use conditions by executing the flow from S210 to S280 in FIG.

[0077] First, in S210, the search unit 140 inputs conditions (also referred to as "fixed conditions") that are not to be searched. For example, in this embodiment, the conditions (also referred to as "search conditions") that are to be searched may be mold conditions and molding conditions. In this case, conditions other than the mold conditions and molding conditions (for example, raw material information and design information) are input as fixed conditions.

[0078] The fixed conditions may be input by the user, or alternatively, predetermined conditions may be input. In S210, the search unit 140 may input, in addition to the fixed conditions, information related to the search, such as criteria used in S260 (described later), a search range of the search conditions, and hyperparameters used in the search.

[0079] Next, in S220, the forming prediction unit 120 generates coordinate information based on the design information. The coordinate information is required for predicting physical properties later, and may be generated in advance by the forming prediction unit 120.

[0080] The coordinate information may be information obtained by dividing the design shape of the foamed resin molded body shown in the design information into a large number of regions of predetermined shapes (hereinafter also referred to as "unit regions"). For example, the design shape may be divided into a cubic grid or rectangular parallelepiped grid of a predetermined size, and the coordinate information may be information regarding the coordinates of the divided grid. For example, the coordinate information may be information regarding the coordinates indicating each unit region, or information regarding the coordinates indicating the division plane, etc.

[0081] The multiple unit areas may be adjacent to each other but not overlap each other. Alternatively, the multiple unit areas may partially overlap each other. The multiple unit areas may include equal amounts of vent holes, core vents, and / or feeder attachment ports. For example, the unit areas may be arranged so that no single unit area includes multiple vent holes, core vents, or feeder attachment ports.

[0082] The size of the cubic lattice or rectangular lattice may be determined based on mold conditions. For example, the size of the cubic lattice or rectangular lattice may be 1 / 10 or 1 / 20 of the size of the air holes, core vents, and / or feeder attachment ports. The molding prediction unit 120 may generate coordinate information based on design information and mold information.

[0083] Next, in S230, the search unit 140 generates search conditions. In the flow of Fig. 3, the mold conditions and molding conditions are the search conditions, so the search unit 140 generates the mold conditions and molding conditions.

[0084] For example, the search unit 140 may generate data on (1) raw material filling time, (2) filling air pressure, (3) heating temperature, (4) heating time, (5) steam pressure, (6) cooling temperature, (7) cooling time, and (8) cracking amount as molding conditions. For example, the search unit 140 may generate data on (1) the number, (2) center coordinate positions, (3) shape (e.g., perfect circle or rectangle), and (4) size (e.g., diameter or long side) of vent holes, core vents, and feeder attachment ports as mold conditions.

[0085] In the first S230, the search unit 140 may set predetermined initial conditions as search conditions, or may set search conditions randomly. In the second or subsequent S230, the search unit 140 may modify the search conditions previously set. For example, the search unit 140 may set search conditions (e.g., mold conditions and molding conditions) that are likely to obtain desirable physical properties using the physical properties acquired in S250, which will be described later. For example, the search unit 140 may modify the mold conditions and molding conditions from the previous conditions using a known search method, such as a gradient method.

[0086] The search unit 140 may set search conditions based on a preset search range. For example, the search unit 140 may set mold conditions and molding conditions so as not to deviate from the search range.

[0087] In S240, the molding prediction unit 120 generates a feature vector based on the mold conditions generated in S230 and the coordinate information generated in S220. The feature vector may include one or more feature amounts related to the positional relationship with the mold for each divided region of the foamed resin molded body defined by the coordinate information.

[0088] For example, the feature vector may include feature quantities related to the positional relationship of each region with the vent holes, core vents, and / or feeder attachment ports of the mold when the foamed resin molded article is placed in the mold. The feature quantities related to the positional relationship with the vent holes, core vents, and / or feeder attachment ports may include the number, orientation, and / or distance of the vent holes, core vents, and / or feeder attachment ports that are present within a predetermined distance from the region. The feature quantities related to the positional relationship with the vent holes, core vents, and / or feeder attachment ports may include the number, orientation, and / or distance of the vent holes, core vents, and / or feeder attachment ports that are nearest to the region.

[0089] That is, the feature vector includes feature amounts generated for multiple regions, and each region includes one or more feature amounts. The forming prediction unit 120 may generate the feature vector by performing coordinate calculations or the like based on the mold conditions and coordinate information.

[0090] In S250, the molding prediction unit 120 predicts the physical properties of the foamed resin molded body for each region based on the raw material information input in S210, the molding conditions generated in S230, and the feature vector generated in S240. For example, the molding prediction unit 120 inputs the raw material information, molding conditions, and feature vector into the molding prediction model trained in S100, and causes the molding prediction model to output the physical properties of the foamed resin molded body for each region.

[0091] In S255, the physical property evaluation unit 150 generates an evaluation value of the physical property of the entire foamed resin molded body based on the physical property of each region predicted in S250. The physical property evaluation unit 150 may generate the evaluation value by calculating the physical property predicted in S250 using a predetermined method.

[0092] For example, the physical property evaluation unit 150 may calculate the sum, average, median, minimum, maximum, variance, and / or standard deviation of the physical properties for each region predicted in S250 as the evaluation value. For example, the physical property evaluation unit 150 may input the physical properties for each region predicted in S250 into a predetermined function and use the output value of the function as the evaluation value. The physical property evaluation unit 150 may calculate the evaluation value after excluding outliers (e.g., a predetermined number of top and bottom values) of the predicted physical properties for each region.

[0093] The physical property evaluation unit 150 may calculate one or more evaluation values. The process of S255 may be omitted. In this case, the physical property evaluation unit 150 may use the physical property itself predicted for each region in S250 as the evaluation value.

[0094] In S260, the physical property evaluation unit 150 determines whether the physical properties predicted in S250 satisfy predetermined criteria. For example, the physical property evaluation unit 150 determines whether the evaluation values ​​calculated in S255 satisfy predetermined criteria. As an example, the physical property evaluation unit 150 may determine whether all or part of the evaluation values ​​are equal to or less than a threshold value or equal to or greater than a threshold value.

[0095] When the physical properties for each region predicted in S250 are used as the evaluation values, the predetermined criteria may be related to the density, compressive strength, tensile strength, flexural strength, water absorption, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and / or dielectric constant of the foamed resin molded body.

[0096] If it is determined that the criteria are met, the property evaluation unit 150 advances the process to S280. If not, the property evaluation unit 150 returns the process to S230 and causes the search unit 140 to generate search conditions again.

[0097] In S280, the proposing unit 170 outputs the search conditions (e.g., mold conditions and molding conditions) that were determined by the physical property evaluating unit 150 in S260 to satisfy the predetermined criteria as proposed conditions (e.g., proposed mold conditions and proposed molding conditions).

[0098] In the above, an example has been described in which the process proceeds from S260 to S280 when the criterion is met for the first time, but this is not limiting. For example, the physical property evaluation unit 150 may return the process to S230 until it determines in S260 that the criterion is met a predetermined number of times, and proceed to S280 only when the criterion is met a predetermined number of times. In S280, the proposal unit 170 may output all of the proposed conditions that meet the criterion, or a portion of the proposed conditions that meet the criterion and have the highest evaluation.

[0099] Fig. 7 shows an example of processing by the shaping prediction unit 120 in the subflow of S200 in Fig. 6. The shaping prediction unit 120 may include a coordinate determination unit 122, a feature vector generation unit 124, and a prediction generation unit 126.

[0100] The coordinate determination unit 122 receives the design information 58 and, if necessary, the mold conditions 54 , generates coordinate information 70 , and outputs it to the feature vector generation unit 124 .

[0101] The feature vector generation unit 124 receives the mold conditions 54 and the coordinate information 70 , generates a feature vector 72 , and outputs it to the prediction generation unit 126 .

[0102] The prediction generation unit 126 inputs the molding conditions 52, raw material information 56, and feature vector 72 into the molding prediction model, and outputs the physical properties 59. In this way, the mold conditions 54 and design information 58 may be converted into the feature vector 72 and then input into the molding prediction model.

[0103] In this way, the foamed resin molded body is divided into a plurality of regions, and a different feature vector 72 is output for each region by the feature vector generation unit 124. As a result, the prediction generation unit 126 predicts and outputs the physical properties of each region of the foamed resin molded body.

[0104] 8A shows an example of the feature vector generated in S240. The many squares surrounded by dashed lines represent the many regions defined by the coordinate information. V01 to V67 represent feature amounts generated by the feature vector generation unit 124 corresponding to each region. The feature vector generation unit 124 may output a vector (V01, V02, V03, ... V67) concatenating the feature amounts V01 to V67 as the feature vector 72 of the foamed resin molded body.

[0105] V01 to V67 may include multiple feature amounts, in which case they are essentially treated as a vector. In this case, the feature vector generation unit 124 may output a vector (V01, V02, V03, ... V67) obtained by connecting the vectors V01 to V67 as the feature vector 72 of the foamed resin molded body.

[0106] FIG. 8B shows an example of physical properties predicted in S250. The many squares surrounded by dashed lines represent many regions defined by coordinate information, and are the same as the regions defined by the feature vectors in FIG. 8A. The molding prediction unit 120 predicts physical properties for each region of the foamed resin molded body. Two types, "good" and "bad," are shown in FIG. 8B. For example, regions with physical properties above a predetermined standard may be considered "good," and other regions may be considered "bad." In this way, according to this embodiment, physical properties can be predicted for each region of the foamed resin molded body.

[0107] After the process of S200 described with reference to FIGS. 3 to 8, the process of S300 is performed. In S300, a foamed resin molded body is actually molded under the proposed conditions. For example, a foamed resin molded body is molded based on the proposed molding conditions using a mold with the proposed mold conditions output in S280. The raw materials, design drawings, etc. may be those based on the raw material information 56 and design information 58. The molding process may be the same as that described with reference to FIGS. 1A to 1D.

[0108] Next, in S400, the physical properties of the foamed resin molded article molded in S300 are acquired. The acquired physical properties may be the same as the physical properties output by the molding prediction model. The physical properties may be acquired by measuring the molded foamed resin molded article using a measurement method and / or measurement device corresponding to the acquired physical properties. For example, the density of the foamed resin molded article may be measured using Archimedes' principle.

[0109] Instead of actually measuring, the surface prediction unit 130 may use a surface prediction model to predict and acquire physical properties from a photographed image of an actually molded foamed resin molded body.

[0110] 9 shows an example of a subflow of acquiring physical properties by the surface prediction unit 130. The surface prediction unit 130 may acquire the physical properties of the foamed resin molded body by executing the flow from S410 to S430 in FIG.

[0111] In S410, the surface prediction unit 130 acquires a captured image of the surface of the foamed resin molded body molded in S300. For example, the captured image may be an image of the foamed resin molded body taken with a camera or an image of the foamed resin molded body scanned with a scanner. The surface prediction unit 130 may acquire the captured image from the camera or scanner.

[0112] Next, in S415, the surface prediction unit 130 may perform image processing on the captured image acquired in S410. For example, the surface prediction unit 130 may convert the captured image into a luminance image (i.e., a grayscale image) based on luminance values. For example, the surface prediction unit 130 may convert the captured image or the luminance image into a binarized image binarized based on luminance values. For example, the surface prediction unit 130 may generate a binarized image by assigning a value of 1 to pixels with luminance greater than a threshold and a value of 0 to pixels with luminance equal to or less than the threshold.

[0113] Next, in S420, the surface prediction section 130 may divide the captured image processed in S415 into a plurality of regions, thereby generating a captured image for each region.

[0114] The regions of the captured image may be adjacent to each other but not overlap each other, or alternatively, the regions of the captured image may partially overlap each other.

[0115] Next, in S430, the surface prediction unit 130 predicts the physical properties of each region of the foamed resin molded body based on the surface prediction model learned in S100. For example, the surface prediction unit 130 may input the captured images of each region obtained in S420 into the surface prediction model and output the physical properties of each region of the foamed resin molded body.

[0116] Fig. 10 shows an example of processing by the surface prediction unit 130 in the subflow of S400 in Fig. 9. The surface prediction unit 130 may include an image processing unit 132 and a surface prediction generation unit 134.

[0117] In S415, the image processing unit 132 processes the captured image 62. In S420, the image processing unit 132 further divides the captured image 62 into regions. The image processing unit 132 outputs a processed image 63 obtained by processing / dividing the captured image.

[0118] The surface prediction generation unit 134 predicts the physical properties in S430. The surface prediction generation unit 134 inputs the processed image 63 into a surface prediction model and outputs the physical properties 64 for each region.

[0119] FIG. 11 shows an example of the captured image processed in S415. The image processing unit 132 may process, for example, a color image of the foamed resin molded body to generate a binarized image as shown. Here, the binarized image may include many regions, each with a different shade of gray. For example, region 90 has a relatively large number of black regions, while region 92 has a relatively small number of black regions. The surface prediction model can predict the physical properties of each region of the foamed resin molded body from the shades of gray in such a binarized image.

[0120] After the process of S400 described with reference to FIGS. 9 to 11, the process of S450 is carried out.

[0121] In S450, the physical property evaluation unit 150 generates an evaluation value of the physical property of the entire foamed resin molded body based on the physical property of each region acquired in S400. The physical property evaluation unit 150 may generate the evaluation value in the same manner as in S255. The evaluation value of the physical property may be the physical property itself.

[0122] In S500, the physical property evaluation unit 150 determines whether the physical properties acquired in S400 satisfy predetermined criteria. For example, the physical property evaluation unit 150 determines whether the evaluation values ​​calculated in S450 satisfy predetermined criteria. As an example, the physical property evaluation unit 150 may determine whether all or part of the evaluation values ​​are equal to or less than a threshold value or equal to or greater than a threshold value.

[0123] The predetermined criteria may be the same as or similar to those used in S260. The same criteria may mean that the types of parameters serving as the criteria are the same but the reference values ​​are different. The predetermined criteria may be different from those used in S260.

[0124] If it is determined that the criteria are met, the physical property evaluation unit 150 may end the process. If not, the physical property evaluation unit 150 may proceed to S600.

[0125] In S600, the learning unit 110 re-learns the prediction model. In the re-learning, the learning unit 110 may use additional training data to train the shaping prediction model. The learning unit 110 may use only the additional training data, or the training data used in S100 in addition to the additional training data to train the shaping prediction model.

[0126] The additional learning data includes pairs of proposed conditions (e.g., mold conditions and / or molding conditions) obtained by the search of the search unit 140 in S200 and physical properties obtained in S400 (i.e., physical properties related to the foamed resin molded body actually molded using the proposed conditions).

[0127] If molding is performed under proposed conditions that are judged to satisfy the criteria in S260, but are judged not to satisfy the criteria in S500, the accuracy of the molding prediction model may be insufficient. Therefore, by re-learning in this embodiment, the prediction system can further improve the prediction accuracy of the molding prediction model. This effect is particularly noticeable when the physical properties of the foamed resin molded product are actually measured in S400.

[0128] In addition to or instead of training the forming prediction model, the training unit 110 may train the surface prediction model using additional training data. If the predicted values ​​of physical properties by the surface prediction unit 130 were obtained in S400, it is possible that the surface prediction model was determined not to satisfy the criteria in S500 due to insufficient accuracy.

[0129] Therefore, the actual physical properties of the foamed resin molded body molded in S300 may be measured, and further, a photographed image of the surface of the foamed resin molded body may be acquired. The learning unit 110 may re-learn the surface prediction model using both as additional learning data.

[0130] After the re-learning in S600, the search for the search conditions is executed again in S200. The forming prediction unit 120 predicts the physical properties again using the forming prediction model re-learned by the learning unit 110. This makes it possible to improve the accuracy of the search after the re-learning. Note that the re-learning is not an essential configuration. In this embodiment, the processes related to the re-learning in S500 to S600 may be omitted.

[0131] As described above, according to this embodiment, it is possible to predict the physical properties of a foamed resin molded product based on mold conditions, molding conditions, etc. Furthermore, it is possible to propose optimal mold conditions, molding conditions, etc. using the prediction results. This makes it possible to maintain the quality of the foamed resin molded product regardless of the mold conditions and molding conditions.

[0132] Furthermore, according to this embodiment, some search conditions that are relatively easily changed, such as mold conditions and molding conditions, are separated from fixed conditions, and a search is performed for only the search conditions. Therefore, according to this embodiment, it is possible to save the computational resources, storage resources, time, etc. required for prediction and recommendation compared to when all conditions are searched using a computer.

[0133] Furthermore, according to this embodiment, the physical properties of a foamed resin molded body can be predicted from a photographed image of the foamed resin molded body. This makes it possible to save resources and time required for measuring physical properties. By using the results of predicting physical properties from such images to retrain the molding prediction model, the accuracy of the molding prediction model can be improved in a short period of time.

[0134] In the above embodiment, the mold conditions and molding conditions are search conditions, and the prediction system 100 searches for mold conditions and molding conditions to be recommended. However, the search targets are not limited to these.

[0135] For example, the prediction system 100 may use either the mold conditions or the molding conditions, or part of the mold conditions and / or the molding conditions, as search conditions. The prediction system 100 may use all or part of the raw material information and / or design information as search conditions in addition to / instead of the mold conditions and molding conditions. This allows the user to search for not only conditions such as mold conditions that are easy to adjust after the fact, but also other conditions as needed.

[0136] Various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where the blocks may represent (1) stages of a process in which operations are performed or (2) sections of an apparatus responsible for performing the operations. Particular stages and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable medium, and / or a processor provided with computer-readable instructions stored on a computer-readable medium. Dedicated circuitry may include digital and / or analog hardware circuitry, and may include integrated circuits (ICs) and / or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry, including logical AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, memory elements such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0137] A computer-readable medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that the computer-readable medium having instructions stored thereon comprises an article of manufacture containing instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable media may include electronic, magnetic, optical, electromagnetic, and semiconductor storage media. More specific examples of computer-readable media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc, memory stick, integrated circuit card, and the like.

[0138] The computer readable instructions may include either assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0139] The computer-readable instructions may be provided to a processor or programmable circuitry of a programmable data processing apparatus, such as a general-purpose computer, special-purpose computer, or other computer, either locally or over a wide-area network (WAN) such as a local area network (LAN), the Internet, etc., which executes the computer-readable instructions to create means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0140] 12 illustrates an example of a computer 2200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and / or to perform a process or steps of a process according to embodiments of the present invention. Such programs may be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.

[0141] A computer 2200 according to this embodiment includes a CPU 2212, a RAM 2214, a graphics controller 2216, and a display device 2218, which are interconnected by a host controller 2210. The computer 2200 also includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the host controller 2210 via an input / output controller 2220. The computer also includes legacy input / output units such as a ROM 2230 and a keyboard 2242, which are connected to the input / output controller 2220 via an input / output chip 2240.

[0142] The CPU 2212 operates according to programs stored in the ROM 2230 and RAM 2214, thereby controlling each unit. The graphics controller 2216 acquires image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214 or into the graphics controller 2216 itself, and causes the image data to be displayed on the display device 2218.

[0143] The communications interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0144] The ROM 2230 stores therein a boot program or the like that is executed by the computer 2200 upon activation, and / or programs that depend on the hardware of the computer 2200. The input / output chip 2240 may also connect various input / output units to the input / output controller 2220 via a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0145] The programs are provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed in the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. Information processing described in these programs is read by the computer 2200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by realizing information manipulation or processing in accordance with the use of the computer 2200.

[0146] For example, when communication is performed between the computer 2200 and an external device, the CPU 2212 may execute a communication program loaded into the RAM 2214 and instruct the communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 2212, the communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in the RAM 2214, the hard disk drive 2224, the DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0147] The CPU 2212 may cause all or a necessary portion of a file or database stored on an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. to be read into the RAM 2214, and may perform various types of processing on the data on the RAM 2214. The CPU 2212 then writes back the processed data to the external recording medium.

[0148] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. The CPU 2212 may perform various types of processing on data read from the RAM 2214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to the RAM 2214. The CPU 2212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, the CPU 2212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0149] The above-described programs or software modules may be stored in a computer-readable medium on or near the computer 2200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable medium, thereby providing the programs to the computer 2200 via the network.

[0150] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0151] It should be noted that the execution order of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before" or "prior to," and that any order may be used unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the process must be performed in that order. The expression "A and / or B" may mean "A, B, or A and C." The expression "A, B, and / or C" may mean "any one of A, B, and C, or any combination of two or more of these."

[0152] In this specification, data may be converted according to a certain rule or formula as necessary. For example, "data of A" does not only refer to "a numerical value or text representing A" itself, but also includes "a numerical value or text representing A" converted according to a certain rule. As an example, if "24°C" and "36°C" exist as data for "temperature (Celsius)," "data for temperature (Celsius)" may include not only "24" and "36," but also "297" (24 + 273) and "309" (36 + 273). [Explanation of symbols]

[0153] 12 resin beads 14 Feeder 22 Mold 24 Mold 30 space 32 Foamed resin molding 50 Forming prediction model 52 Molding conditions 54 Mold conditions 56 Raw material information 58 Design information 59 Physical Properties 60 Surface Prediction Model 62 captured images 63 processed images 64 Physical Properties 70 Coordinate Information 72 feature vectors 90 areas 92 areas 100 Prediction System 110 Learning Department 120 Forming Prediction Department 122 Coordinate determination unit 124 Feature Vector Generation Unit 126 Prediction Generation Unit 130 Surface Prediction Unit 132 Image processing section 134 Surface Prediction Generation Unit 140 Search Department 150 Physical Properties Evaluation Department 170 Proposal Department 2200 Computer 2201 DVD-ROM 2210 host controller 2212 CPU 2214 RAM 2216 Graphics Controller 2218 Display Device 2220 Input / Output Controller 2222 communication interface 2224 hard disk drive 2226 DVD-ROM drive 2230 ROM 2240 I / O chip 2242 keyboard

Claims

1. A prediction system for predicting physical properties of a foamed resin molded body molded in a mold, comprising: a molding prediction unit that inputs at least one of mold conditions related to the connection between the inside and outside of the mold used for molding and molding conditions related to the conditions applied in the molding process, and predicts the physical properties of the foamed resin molded body molded in the mold; Prediction system.

2. the molding prediction unit inputs both the mold conditions and the molding conditions to predict the physical properties of the foamed resin molded body; The prediction system of claim 1 .

3. The mold conditions include at least one of ventilation conditions, which are the number, position, shape, and size of vent holes and core vents that connect the inside and outside of the mold and allow gas to pass through, and feeder attachment conditions, which are the number, position, shape, and size of feeder attachment ports that supply raw materials for the foamed resin molded body to the mold. The prediction system of claim 1 .

4. The molding conditions include at least one of information on a heating temperature, a heating time, a vapor pressure, a cooling temperature, a cooling time, a raw material filling time, a filling air pressure, and a cracking amount. The prediction system of claim 1 .

5. the molding prediction unit further inputs raw material information regarding the raw materials of the foamed resin molded body to predict the physical properties of the foamed resin molded body. The prediction system of claim 1 .

6. the molding prediction unit further inputs design information relating to a design shape of the foamed resin molded body and predicts the physical properties of the foamed resin molded body. The prediction system of claim 1 .

7. The molding prediction unit predicts one or more of the density, compressive strength, tensile strength, flexural strength, water absorption rate, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body as the physical properties of the foamed resin molded body. The prediction system of claim 1 .

8. The molding prediction unit sets a plurality of regions into which the foamed resin molded body is divided, and predicts the physical properties of the foamed resin molded body for each of the regions. The prediction system of claim 1 .

9. The method further includes a physical property evaluation unit that determines whether the predicted physical properties of the foamed resin molded body exceed a predetermined standard. The prediction system of claim 1 .

10. The apparatus further includes a search unit that searches for mold conditions and / or molding conditions that satisfy a predetermined criterion using the prediction result of the molding prediction unit. The prediction system of claim 1 .

11. a learning unit that re-learns a molding prediction model used for prediction by the molding prediction unit using learning data including proposed conditions, which are mold conditions and / or molding conditions obtained by the search by the search unit, and the physical properties of a foamed resin molded body actually molded using the proposed conditions, The forming prediction unit predicts the physical properties using the forming prediction model re-trained by the learning unit. The prediction system of claim 10.

12. The method further includes a surface prediction unit that predicts the physical properties of each region of the foamed resin molded body based on a captured image of the surface of the foamed resin molded body, The physical properties included in the learning data are physical properties predicted by the surface prediction unit from a photographed image of the actually molded foamed resin molded body. The prediction system of claim 11.

13. A surface prediction unit that predicts physical properties of each region of the foamed resin molded body based on a captured image of the surface of the foamed resin molded body, Prediction system.

14. The surface prediction unit predicts one or more of the density, compressive strength, tensile strength, flexural strength, water absorption rate, flexural elasticity, flammability, thermal dimensional change rate, thermal conductivity, and dielectric constant of the foamed resin molded body as the physical properties of the foamed resin molded body. The prediction system of claim 13.

15. the surface prediction unit predicts the physical properties based on a photographed image of the foamed resin molded body taken using reflected light and / or transmitted light. The prediction system of claim 13.

16. the surface prediction unit predicts the physical properties based on a luminance image of the foamed resin molded body. The prediction system of claim 13.

17. the surface prediction unit predicts the physical properties using a binarized image obtained by binarizing the luminance image. The prediction system of claim 16.

18. The prediction system according to any one of claims 1 to 12, a molding prediction step of inputting at least one of the mold conditions and the molding conditions to predict the physical properties of the foamed resin molded body; A prediction method comprising:

19. The prediction system according to any one of claims 13 to 17, a surface prediction step of predicting physical properties of each region of the foamed resin molded body based on a photographed image of the surface of the foamed resin molded body; A prediction method comprising:

20. The method is executed by a computer, causing the computer to: The prediction system according to any one of claims 1 to 17 is operated as follows: program.

Images