JP7765175B2 - Physical property prediction system, physical property prediction device, and physical property prediction method - Google Patents

Description

The present invention relates to a property prediction system, a property prediction device, and a property prediction method for predicting the properties of a composition composed of multiple materials.

A prediction method for determining specific performance of rubber-like elastomers, which are materials used in automobile tires and other applications, is known (see Patent Document 1). This prediction method determines a specific performance (first performance) of the rubber-like elastomer using a relational equation that shows the relationship between molecular information (first data) about individual materials containing multiple polymers and multiple additives, the material blending ratios of multiple types of rubber-like elastomers after vulcanization (second data), vulcanization conditions (third data), and information about the first performance to be predicted (fourth data).

JP 2018-147460 A

Conventional prediction methods allow for the advance prediction of specific performance values for a rubber-like elastomer based on various information about the target rubber-like elastomer. Therefore, developers of rubber-like elastomers can obtain a rubber-like elastomer composition with desired performance values without actually manufacturing the rubber-like elastomer by repeatedly inputting composition information, including the combination and distribution amounts of multiple materials that make up the rubber-like elastomer, to calculate the predicted value. This can speed up the development of rubber-like elastomers and reduce development costs. However, conventional prediction methods do not allow for the understanding of the relationship between the characteristic values of any of the materials in the target rubber-like elastomer and the predicted value. If developers could understand this relationship, they could use that relationship as a starting point to more quickly identify a rubber-like elastomer composition with the desired performance value, which is believed to contribute to the development of rubber-like elastomers.

The object of the present invention is to provide a property prediction system, property prediction device, and property prediction method that can not only predict the properties of a composition made up of multiple materials, but also display the predicted values in association with the characteristic quantities of specific materials contained in the composition.

(1) A physical property prediction system according to one aspect of the present invention is a physical property prediction system that predicts the physical properties of a composition composed of multiple materials. The physical property prediction system includes a material information storage unit that stores material information related to the multiple materials, the material information including feature quantities of a specific material among the multiple materials; a prediction unit that predicts the value of the physical property based on the material information stored in the material information storage unit; and a display processing unit that displays the predicted value predicted by the prediction unit in association with the feature quantities stored in the material information storage unit.

This configuration enables the display processing unit to display the predicted values of the physical properties of the composition on a predetermined display unit. As a result, if the user is a developer of the composition, for example, the user can easily grasp at a glance the relationship between the characteristic quantities of the specific materials that make up the composition and the predicted values of the physical properties of the composition. As a result, the user can use this relationship as a starting point to more quickly find the composition that has the desired physical property values, which can greatly contribute to the development of the composition. Examples of such compositions include a polymer composition containing a polymer made up of multiple materials, and a rubber material made from the polymer composition.

(2) The physical property prediction system of the present invention further includes a feature quantity change unit that changes the feature quantity displayed on the predetermined display unit by the display processing unit. When the feature quantity is changed by the feature quantity change unit, the prediction unit re-predicts the value of the physical property based on the material information including the changed feature quantity.

With this configuration, the physical property values of the composition are re-predicted based on the material information including the changed feature values, and the re-predicted predicted values are displayed on a specified display unit. This allows the user to easily understand the relationship between the changed feature values and the re-predicted predicted values.

(3) The physical property prediction system of the present invention further includes a first selection unit that selects the physical property for which the display processing unit displays the predicted value. In this case, the display processing unit displays the predicted value of the physical property selected by the first selection unit in association with the feature amount.

With this configuration, the user can specify any physical property they wish to have the prediction unit predict.

(4) The physical property prediction system of the present invention further includes a second selection unit that selects the specific material for which the feature is to be displayed in the display processing unit. In this case, the display processing unit displays the predicted value predicted by the prediction unit in association with the feature of the specific material selected by the second selection unit.

With this configuration, the user can specify any material as the specific material to be displayed in association with the predicted value.

(5) The display processing unit displays, on a predetermined display unit, a characteristics graph showing the predicted values corresponding to the possible quantitative range of the feature of the specific material selected by the second selection unit.

This configuration allows the user to easily grasp at a glance the change trend of the predicted value within the quantitative range.

(6) The physical property prediction system of the present invention further includes a range change unit that changes the possible quantitative range of the feature of the specific material selected by the second selection unit. In this case, the display processing unit displays the characteristic graph corresponding to the quantitative range changed by the range change unit on a predetermined display unit.

With this configuration, the user can easily grasp the change trend of the predicted value in response to changes in the quantitative range by looking at the characteristic graph redisplayed by the display processing unit.

(7) When the prediction unit predicts the values of the physical properties of each of the multiple types of compositions, the display processing unit displays the predicted values corresponding to each composition in association with the feature quantities of the specific material corresponding to each predicted value.

This configuration allows the user to easily grasp at a glance the relationship between the predicted values and the feature quantities for each of the different types of compositions.

(8) The physical property prediction system of the present invention further includes an input receiving unit that receives input of the material information of the composition that is the target of prediction by the prediction unit. In this case, the prediction unit predicts the value of the physical property of the composition based on a prediction model based on training data including feature quantities of multiple material information for each of multiple other compositions and values of the physical properties for each of the multiple other compositions, and on the material information received by the input receiving unit.

(9) It is preferable that the characteristic amount includes at least one of the blend amount of the specific material or the physical property value of a predetermined physical property of the specific material.

(10) The physical properties preferably include at least one of hardness, dynamic viscoelasticity at low and high temperatures, tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300% modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristic value, scorch time, and Mooney viscosity. The dynamic viscoelasticity may include complex modulus E*, storage modulus E', loss modulus E", loss tangent tanδ, complex shear modulus G*, storage shear modulus G', loss shear modulus G", etc.

(11) The composition preferably contains at least a polymer and an additive. In this case, the material information is information about the molecules of the polymer and the additive, and preferably includes at least one of the weight average molecular weight, number average molecular weight, molecular weight distribution, and degree of branching of molecular chains.

(12) It is preferable that the material information includes at least one of the cis-isomer ratio, trans-isomer ratio, oil extension amount, glass transition temperature, solubility parameter, styrene content, vinyl content, butadiene rubber content, and viscoelastic properties of the polymer.

(13) It is preferable that the additive includes a filler, and the material information includes at least one of the particle size, CTAB adsorption specific surface area, BET adsorption specific surface area, and surface polarity of the filler.

(14) A physical property prediction device according to another aspect of the present invention is a physical property prediction device that predicts the physical properties of a composition composed of multiple materials. The physical property prediction device includes a material information storage unit that stores material information related to the multiple materials, the material information including feature quantities of a specific material among the multiple materials; a prediction unit that predicts the value of the physical property based on the material information stored in the material information storage unit; and a display processing unit that displays the predicted value predicted by the prediction unit in association with the feature quantities stored in the material information storage unit.

(15) A physical property prediction method according to another aspect of the present invention is a physical property prediction method for predicting the value of a physical property of a composition composed of multiple materials. The physical property prediction method includes a prediction step for predicting the value of the physical property based on material information about the multiple materials, and a display processing step for displaying the predicted value predicted in the prediction step in association with the feature value of a specific material from the multiple materials.

This invention not only predicts the physical properties of a composition made up of multiple materials, but also makes it possible to correlate and display the predicted values with the characteristic quantities of specific materials contained in the composition.

FIG. 1 is a diagram showing the configuration of a rubber performance prediction system according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of a prediction device included in the rubber performance prediction system. FIG. 3 is a block diagram showing the configuration of an information terminal provided in the rubber performance prediction system. FIG. 4 is a diagram showing an example of an input screen (mixture ratio input screen) displayed on an information terminal used by a user. FIG. 5 is a diagram showing an input operation of a material category on the blending ratio input screen of the information terminal. FIG. 6 is a diagram showing the input operation of materials on the blending ratio input screen of the information terminal. FIG. 7 is a diagram showing the input operation of the mixture ratio on the mixture ratio input screen of the information terminal. FIG. 8 is a diagram showing an input operation for shifting from the blending ratio input screen to the material property input screen of the information terminal. FIG. 9 is a diagram showing the input operation of material properties on the material property input screen of the information terminal. FIG. 10 is a diagram showing the input operation of material properties on the material property input screen of the information terminal. FIG. 11 is a diagram showing an output setting screen displayed on an information terminal used by a user. FIG. 12 is a diagram showing an input operation for predicted performance on the output setting screen of the information terminal. FIG. 13 is a diagram showing a predicted execution operation on the output setting screen of the information terminal. FIG. 14 is a diagram showing an example of a prediction result display screen displayed on an information terminal used by a user. FIG. 15 is a diagram showing a characteristic value changing operation on the prediction result display screen of the information terminal. FIG. 16 is a diagram showing an example of a prediction result display screen after re-prediction on the prediction result display screen of the information terminal. FIG. 17 is a flowchart illustrating an example of a procedure for a performance prediction process executed by the control unit of the prediction device. FIG. 18 is a diagram illustrating an example of a prediction result display screen on which a prediction graph is displayed. FIG. 19 is a diagram showing an example of a prediction result display screen in which a coordinate point on a prediction graph is selected. FIG. 20 is a diagram showing a display frame in which a three-dimensional coordinate system is displayed.

Embodiments of the present invention will now be described with reference to the accompanying drawings. Note that the following embodiments are examples of specific embodiments of the present invention and do not limit the technical scope of the present invention.

Figure 1 is a diagram showing the configuration of a rubber performance prediction system 100 (hereinafter simply referred to as prediction system 100) according to an embodiment of the present invention. The prediction system 100 of this embodiment is composed of a prediction device 10, an information terminal 20, and a database 30. Note that the prediction system 100 is an example of a physical property prediction system of the present invention. Furthermore, the prediction device 10 is an example of a physical property prediction device of the present invention.

Prediction system 100 is a system that predicts specific performance of a rubber-like elastomer (one example of a composition of the present invention) made from a polymer composition composed of multiple materials including at least a polymer and additives. The rubber-like elastomer is obtained by vulcanizing the polymer composition, and specifically, is a rubber material used in the manufacture of tire products such as pneumatic tires mounted on vehicles such as automobiles. In other words, prediction system 100 of this embodiment predicts specific performance that indicates the characteristics of the rubber-like elastomer that makes up a tire product. The prediction results (predicted values) by prediction system 100 are used in the development of the tire product.

While this embodiment uses a rubber material used in the manufacture of the tire product as an example of the rubber-like elastomer, the rubber-like elastomer may also be the tire product itself. In this case, the prediction system 100 predicts a specific performance among the multiple tire performance characteristics of the tire product. The rubber-like elastomer may also be an industrial rubber product itself, such as vibration-damping rubber, or a rubber material used in the manufacture of industrial rubber products. While this embodiment uses a rubber product obtained by vulcanizing the polymer composition as an example, as described below, the prediction system may also predict a specific performance of a polymer composition in an unvulcanized state (i.e., unvulcanized rubber). In this case, the unvulcanized polymer composition (unvulcanized rubber) is an example of the composition of the present invention.

The polymer composition is an unvulcanized rubber made by mixing multiple materials, including one or more polymers and one or more additives. The rubber-like elastomer is made by vulcanizing the polymer composition.

The polymer is, for example, unvulcanized raw rubber that is blended into the polymer composition. Examples of raw rubber include natural rubber (NR), isoprene rubber (IR), butadiene rubber (BR), and styrene-butadiene rubber (SBR).

Such additives include, for example, fillers such as carbon black and silica, antioxidants, vulcanization accelerators, oil, zinc oxide, stearic acid, sulfur, and processing aids.

[Configuration of prediction system 100]
1, the prediction system 100 includes a prediction device 10, an information terminal 20, and a database 30, which are communicably connected to each other via a network N1. The network N1 is, for example, a wired communication network connected by a LAN or the like, or a wireless communication network such as a dedicated line or a public line.

In this embodiment, a configuration in which the database 30 is connected to the network N1 is illustrated, but the database 30 may be provided in the prediction device 10 or the information terminal 20, for example. In addition, in this embodiment, a configuration in which the information terminal 20 is connected to the network N1 is illustrated, but the various components and functions provided in the information terminal 20 may be installed in the prediction device 10.

The prediction device 10 is one element constituting the prediction system 100. Using input information transmitted from the information terminal 20 and a pre-constructed prediction model 123 (see FIG. 2 ), the prediction device 10 predicts the specific performance of the rubber-like elastic material to be predicted (hereinafter, the rubber-like elastic material to be predicted may be referred to as the "predicted rubber") and outputs the prediction results to the information terminal 20. The prediction device 10 is an information processing device capable of executing various types of arithmetic processing, such as a server computer, cloud server, or personal computer connected to the network N1. The prediction device 10 is not limited to a single computer, but may also be a computer system in which multiple computers operate in cooperation, or a cloud computing system. The various processes executed by the prediction device 10 may be executed in a distributed manner by one or multiple processors. A program or computer software for operating the prediction system 100 is installed in the prediction device 10.

The prediction device 10 constructs a prediction model 123 (see Figure 2) based on training data stored in the database 30, and uses the prediction model 123 to predict the specific performance of the rubber to be predicted. In this embodiment, the prediction device 10 predicts the specific performance of the rubber-like elastic body used in the tire product, and the prediction result is evaluated as the predicted value of the specific performance of the rubber to be predicted, or as the predicted value of the specific performance of the tire product. The prediction device 10 and the prediction model 123 will be described later.

The information terminal 20 is an information processing device or terminal device used by a user. The information terminal 20 is a desktop personal computer, a laptop computer, or a portable terminal such as a smartphone or tablet. The user inputs various information about the predicted rubber through the information terminal 20. The information terminal 20 also displays the prediction results sent from the prediction device 10 on its display screen. Therefore, a program or computer software is installed on the information terminal 20 to work in conjunction with the prediction system 100 to send the various information to the prediction device 10 and display the prediction results on its display screen.

Database 30 is a collection of data in which various data handled by prediction system 100 is stored on a storage medium based on a predetermined data management method. Database 30 is managed in various forms, such as a storage device, information processing device, cloud server, or data server communicatively connected to network N1. Database 30 includes training data for generating a prediction model 123 (see Figure 2) used in prediction processing by prediction device 10.

The training data is information used to generate the prediction model 123. Specifically, the training data includes material information for numerous sample rubbers Tk (k = 1, 2, ..., n) consisting of numerous rubber-like elastomers with known performance characteristics, and physical property values (so-called training information) for the various performance characteristics of each sample rubber Tk. The sample rubbers Tk are rubber products (tire products, industrial rubber products, etc.) previously manufactured as finished products, prototype rubber products manufactured during research for the development of the rubber products, or test pieces consisting of the rubber-like elastomers manufactured during the research. A dataset including the material information and physical property values for each of the numerous sample rubbers Tk is stored in the database 30 as the training data. The sample rubbers Tk are an example of the multiple other compositions of the present invention.

The material information includes, for example, information on the properties of each material, such as the polymer Pk (k = 1, 2, ..., n) and additives that make up the sample rubber Tk (hereinafter referred to as material property information), the compounding ratio of each material in the sample rubber Tk, etc.

Here, the material property information of the polymer is preferably at least one of the weight average molecular weight (Mw), number average molecular weight (Mn), molecular weight distribution (Mw/Mn), molecular chain branching degree (polymer linearity), or peak top molecular weight (Mp). The material property information of the polymer may also include at least one of the isoprene rubber amount, styrene amount, vinyl amount, butadiene rubber amount, trans-butadiene rubber amount ratio, or cis-butadiene rubber amount ratio. In this embodiment, the weight average molecular weight (Mw), number average molecular weight (Mn), molecular weight distribution (Mw/Mn), isoprene rubber amount, styrene amount, vinyl amount, trans-butadiene rubber amount, and cis-butadiene rubber amount, which are exemplified as the material property information of the polymer, are included in the training data. Table 1 shows an example of the material property information for various polymers Pk (k = 1, 2, ..., n) included in the training data. The material property information of the polymer Pk may include at least one of the amount of oil extension, glass transition temperature, solubility parameter, viscoelastic properties, type of modifying group, and viscosity.

Furthermore, when the additive is a filler such as carbon black or silica, the material property information of the additive preferably includes at least one of the filler type (carbon black, silica, etc.), particle size (primary particle size), CTAB adsorption specific surface area, BET adsorption specific surface area, or surface polarity.

Table 2 shows an example of the blending ratio of each material in the sample rubber Tk (T1, T2, ..., Tn). In Table 2, the blending ratio of the materials (polymers, additives, etc.) in each sample rubber Tk is expressed in parts by mass, and for each additive Ak (k = 1, 2, ..., n), the blending ratio is shown when the total value of each polymer is 100 parts by mass.

The various performance (physical properties) of the sample rubber Tk preferably include at least one of the following: hardness, dynamic viscoelasticity at low and high temperatures, tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300% modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristic values, scorch time, and Mooney viscosity. The vulcanization characteristic values are measured using a specified vulcanization tester (curelastometer), such as induction time tC(10), 50% vulcanization time tC(50), and 90% vulcanization time tC(90). The scorch time and Mooney viscosity are measured using a specified Mooney viscometer. The dynamic viscoelasticity includes the complex modulus E*, storage modulus E', loss modulus E", loss tangent tanδ, complex shear modulus G*, storage shear modulus G', and loss shear modulus G". In this embodiment, the training data includes the physical property values of all of the above-mentioned performances exemplified as various performances of the sample rubber Tk. Table 3 shows examples of the physical property values (measured values) of various performances of each of the sample rubbers Tk (T1, T2, ..., Tn).

The material property information (see Table 1) included in the training data, the information on the physical properties of the sample rubber Tk (see Table 3), and the methods for measuring this information are well known and are also described in detail in the above-mentioned prior art document (JP 2018-147460 A), so a detailed explanation will be omitted here.

In this embodiment, as described above, the training data includes the material information and the physical property values. However, the training data is not limited to the examples described in this embodiment. For example, the material information may include the mixing conditions for mixing the polymer composition. This is because the mixing conditions may affect the physical property values of the rubber to be predicted. The mixing conditions may include at least one of the following: the volume of the chamber of the mixer, the amount of polymer composition filled during mixing, the mixing time, and the discharge temperature of the polymer composition after mixing. The material information may also include vulcanization conditions for vulcanizing the polymer composition. This is because the vulcanization conditions may affect the physical property values of the rubber to be predicted. The vulcanization conditions may be, for example, the vulcanization temperature conditions set for the polymer composition before vulcanization of each sample rubber Tk. Note that the mixing conditions and vulcanization conditions are well known and are described in detail in the aforementioned prior art document (JP 2018-147460 A), so a detailed description thereof will be omitted here.

[Prediction device 10]
The configuration of the prediction device 10 will be described below with reference to Fig. 2. Fig. 2 is a block diagram showing the configuration of the prediction device 10.

The prediction device 10 is used to realize the prediction system 100 of this embodiment, and as shown in FIG. 2, includes a control unit 11, a memory unit 12, a communication unit 13, a display unit 14, and an operation unit 15.

The communication unit 13 is a communication interface that connects the prediction device 10 to the network N1 and performs data communication with each device connected to the network N1 in accordance with a predetermined communication protocol. Specifically, the communication unit 13 performs data communication with the information terminal 20 and the database 30 via the network N1.

The storage unit 12 is a non-volatile storage medium such as an HDD, SSD, or flash memory that stores various types of information and data. The storage unit 12 stores a control program 121 and a prediction model 123. Note that the prediction model 123 may be realized as an electronic circuit that includes a memory in which the prediction model 123 is stored.

The control program 121 may be non-temporarily recorded on a computer-readable recording medium such as a CD or DVD, read from the recording medium by a reading device (not shown) such as a CD drive or DVD drive electrically connected to the prediction device 10, and copied to the memory unit 12. The control program 121 may also be read from external storage connected to the network N1, input via the communication unit 13, and copied to the memory unit 12.

The control program 121 is a program that causes the control unit 11 to execute the prediction process (see Figure 17) described below using the prediction model 123.

Prediction model 123 is a trained model used in the prediction process (see Figure 17) described below, and calculates predicted values for the physical property values of a specific performance of the rubber-like elastic material (predicted rubber) that is the prediction target. In this embodiment, when the compounding ratio of the materials of the rubber-like elastic material to be predicted and the physical property values of the characteristics of each material are input to the input section, prediction model 123 predicts the physical property values of the specific performance of the rubber-like elastic material and outputs the predicted value from the output section. Note that prediction model 123 may include a well-known function (prediction function) that returns a predicted value (output data) for an input value (input data).

The prediction model 123 is generated by machine learning by the control unit 11 based on the training data stored in the database 30 and a predetermined algorithm. Note that the control unit 11 updates the prediction model 123 each time the content of the training data is updated. Note that the prediction model 123 may be one that has been learned and generated by a control unit other than the control unit 11, transferred from an external unit, and stored in the memory unit 12.

In this embodiment, the prediction model 123 can be constructed using various methods. For example, algorithms required to construct the prediction model 123 include multiple regression, generalized linear regression, principal component regression, ridge regression, lasso regression, kernel regression, random forest regression, Gaussian process regression, multilayer neural networks, clustering, support vector machines, and RBF networks defined by radial basis functions. The prediction model 123 may also be constructed using deep learning, which multilayers the intermediate layers of a neural network. The prediction model 123 may use one of the above-mentioned algorithms, or multiple algorithms. In this embodiment, the generalized linear regression, which offers an excellent balance between accuracy and cost, is used as the algorithm for the prediction model 123.

In this embodiment, the input values (design variables) for the prediction model 123 are material information for each material (polymer, additive) contained in the sample rubber Tk, and the output values (objective variables) are physical property values of the various performance characteristics of the sample rubber Tk.

Such a predictive model 123 can be constructed using commercially available computer software (e.g., MATLAB (registered trademark) from The MathWorks, Inc., or modeFRONTIER from ESTECO, Inc.).

The display unit 14 is a display device such as a liquid crystal display or organic EL display that displays various information. The operation unit 15 is an input device such as a mouse, keyboard, or touch panel that accepts operations from the operator.

The control unit 11 controls the operation of each unit of the prediction device 10. The control unit 11 has control devices such as a CPU, ROM, and RAM. The CPU is a processor that executes various types of arithmetic operations. The ROM is a non-volatile storage medium that pre-stores control programs such as BIOS and OS that cause the CPU to execute various types of arithmetic operations. The RAM is a volatile or non-volatile storage medium that stores various types of information and is used as temporary storage memory (work area) for the various types of arithmetic operations executed by the CPU. The control unit 11 controls the prediction device 10 by having the CPU execute various control programs pre-stored in the ROM or memory unit 12.

As shown in FIG. 2, the control unit 11 includes various processing units, such as a performance prediction unit 111 (an example of a prediction unit of the present invention). The control unit 11 functions as the performance prediction unit 111 when the CPU executes various arithmetic processes in accordance with the control program. The control unit 11 or the CPU is an example of a computer that executes the control program. Note that part or all of the performance prediction unit 111 may be configured as an electronic circuit. The control program may also be a program for causing multiple processors to function as the performance prediction unit 111. The control program may also be a program for causing a graphics processing unit (GPU) to function as the performance prediction unit 111.

The performance prediction unit 111 predicts specific performance (specific performance) of the rubber to be predicted based on material information about the multiple materials (polymers, additives, etc.) that make up the rubber to be predicted. The predicted specific performance includes various properties of the rubber to be predicted, such as hardness, dynamic viscoelasticity at low and high temperatures, tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300% modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristics, scorch time, and Mooney viscosity. The vulcanization characteristics are measured using a specified vulcanization tester (curelastometer), such as induction time tC(10), 50% vulcanization time tC(50), and 90% vulcanization time tC(90). The scorch time and Mooney viscosity are measured using a specified Mooney viscometer. Furthermore, the dynamic viscoelasticity includes the complex modulus E*, storage modulus E', loss modulus E", loss tangent tanδ, complex shear modulus G*, storage shear modulus G', and loss shear modulus G".

The material information is input by the user on the information terminal 20, and the input information is transferred to the prediction device 10 via the network N1. The material information of the rubber to be predicted includes the compounding ratio of the materials (polymers, additives, etc.) of the rubber to be predicted, and material property information of the polymers, additives, etc. that make up the rubber to be predicted. The material information transferred to the prediction device 10 is stored in RAM or memory unit 12 within the control unit 11 of the prediction device 10. In this case, the RAM or memory unit 12 is an example of a material information memory unit of the present invention.

When the performance prediction unit 111 acquires the material information of the rubber to be predicted from the information terminal 20, it predicts the specific performance of the rubber to be predicted using the prediction model 123 and the acquired material information. Specifically, the performance prediction unit 111 inputs the material information into the input section of the prediction model 123, causes the prediction model 123 to predict the specific performance of the rubber to be predicted, and acquires the physical quantity indicating the specific performance output from the output section of the prediction model 123 as the prediction result (predicted value).

The predicted values predicted by the performance prediction unit 111 are transferred to the information terminal 20 by the communication unit 13 for display on the display unit 23 of the information terminal 20.

In this embodiment, the specific performance predicted by the performance prediction unit 111 is selected by a first selection processing unit 212 (described below) of the information terminal 20. Selection information regarding the specific performance selected by the first selection processing unit 212 is transferred from the information terminal 20 to the prediction device 10. When the control unit 11 receives the selection information, it sets the performance indicated by the selection information as the target for prediction by the performance prediction unit 111. In this case, the performance prediction unit 111 predicts the specific performance selected by the first selection processing unit 212.

The performance prediction unit 111 is not limited to predicting only the performance selected by the first selection processing unit 212, but may also predict, for example, some or all of the various performance characteristics of the rubber to be predicted (hardness, dynamic viscoelasticity at low and high temperatures, tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300% modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristic values, scorch time, and Mooney viscosity).

When the prediction model 123 makes a prediction using the prediction function, for example, when the material information is x and the predicted value of the specific performance is y, the prediction function is expressed as y = f(x). When the material information x and the specific performance y of each of n sample rubbers Tk exist as training data, the prediction function can be estimated from n paired data D expressed by the following formula:
D = {(x1, y1), (x2, y2), ..., (xN, yN)}

[Information terminal 20]
The configuration of the information terminal 20 will be described below with reference to Fig. 3. Fig. 3 is a block diagram showing the configuration of the information terminal 20.

As shown in FIG. 3, the information terminal 20 includes a control unit 21, a memory unit 22, a display unit 23, an operation unit 24, a communication unit 25, etc. The information terminal 20 is an information processing device or terminal device used by a user. The user inputs various information about the predicted rubber required for prediction processing by the prediction device 10 from the information terminal 20.

The communication unit 25 is a communication interface that connects the information terminal 20 to the network N1 and performs data communication with the prediction device 10 via the network N1 in accordance with a predetermined communication protocol.

The memory unit 22 is a non-volatile storage medium such as a flash memory that stores various types of information. The memory unit 22 stores a control program 221 that causes the control unit 21 to execute various types of calculation processing, as well as data used in the various types of calculation processing.

The display unit 23 is a display device such as a liquid crystal display or organic EL display that displays various information. The operation unit 24 is an input device such as a mouse, keyboard, or touch panel that accepts input operations by the user.

In this embodiment, the predicted value predicted by the performance prediction unit 111 of the prediction device 10 and the characteristic value of a pre-selected specific material (hereinafter referred to as the material characteristic value) are displayed on the display unit 23. Specifically, coordinate points indicating the relationship between the material characteristic value and the predicted value are displayed in a display frame 65 on the prediction result display screen 234 (see FIG. 14) displayed on the display unit 23. The material characteristic value is an example of a feature quantity of the present invention.

The control unit 21 controls the operation of each part of the information terminal 20. The control unit 21 has control devices such as a CPU, ROM, and RAM. The CPU is a processor that executes various types of arithmetic operations. The ROM is a non-volatile storage medium that pre-stores control programs such as BIOS and OS that cause the CPU to execute various types of arithmetic operations. The RAM is a volatile or non-volatile storage medium that stores various types of information, and is used as temporary storage memory (work area) for the various types of arithmetic operations executed by the CPU. The control unit 21 controls the information terminal 20 by having the CPU execute various control programs pre-stored in the ROM or memory unit 22.

As shown in FIG. 3, the control unit 21 includes various processing units, such as an input receiving unit 211 (an example of an input receiving unit of the present invention), a first selection processing unit 212 (an example of a first selection unit of the present invention), a second selection processing unit 213 (an example of a second selection unit of the present invention), a characteristic value changing unit 214 (an example of a feature value changing unit of the present invention), a range changing unit 215 (an example of a range changing unit of the present invention), and a display processing unit 216 (an example of a display processing unit of the present invention). The control unit 21 functions as the various processing units when the CPU executes various arithmetic processes in accordance with the control program. The control unit 21 or the CPU is an example of a computer that executes the control program. Note that some or all of the processing units included in the control unit 21 may be configured as electronic circuits. Furthermore, the control program may be a program for causing multiple processors to function as the various processing units.

The input acceptance unit 211 performs processing to accept input of the material information of the rubber to be predicted, which is the target of prediction by the performance prediction unit 111 of the prediction device 10. Specifically, the input acceptance unit 211 displays multiple input screens 231, 232 on the display unit 23, and acquires information entered into each input screen 231, 232 by the operation unit 24. The acquired information is then transmitted to the prediction device 10 by the communication unit 25.

Figures 4 to 8 show a compounding ratio input screen 231 (an example of an input screen) displayed on the display unit 23 of the information terminal 20. The compounding ratio input screen 231 is an input screen for inputting the compounding ratios of all materials that make up the predicted rubber.

As shown in FIG. 4, the blending ratio input screen 231 includes an input frame 41 for inputting the material category, an input frame 42 for inputting the material, an input frame 43 for inputting the blending ratio, and a display frame 44 for displaying the blending ratio. The blending ratio input screen 231 also has a back key 31 for returning the displayed content to the previous screen, a registration key 32 for confirming the input content, and a next key 33 for moving the displayed content to the next screen. The user can input any information into the input frames 41 to 43 by operating the operation unit 24.

The input box 41 is used to input the category of the material that makes up the predicted rubber. The input box 41 is provided with a pull-down key 41A, and when the pull-down key 41A is pressed using the operation unit 24, a pull-down menu 41B (see Figure 5) containing category information for multiple items that have been registered in advance is displayed.

Figure 5 shows the state when pull-down key 41A is pressed and pull-down menu 41B is opened. As shown in Figure 5, pull-down menu 41B contains the following items as category information: "Polymer," "Filler," "Vulcanization Accelerator," "Coupling Agent," and "Other." When a category of the material that makes up the predicted rubber is selected from these, the selected category is displayed in input frame 41. For example, when pull-down key 41A is pressed and then "Polymer" is selected from pull-down menu 41B, "Polymer" is displayed in input frame 41.

The input frame 42 is used to input a specific material belonging to a specific category. Once the category input in the input frame 41 is complete, the material can be input in the input frame 42. For example, if "polymer" is input in the input frame 41, multiple polymers that can be used as the material for the rubber-like elastic body can be input in the input frame 42. The input frame 42 is provided with a pull-down key 42A, and when the pull-down key 42A is pressed using the operation unit 24, a pull-down menu 42B (see Figure 6) containing multiple pre-registered polymers Pk (P1, P2, ..., Pn) is displayed.

Figure 6 shows the state when the pull-down key 42A is pressed and the pull-down menu 42B is opened. As shown in Figure 6, when a specific polymer that constitutes the predicted rubber is selected from the multiple polymers displayed in the pull-down menu 42B, the selected polymer is displayed in the input frame 42. For example, when the pull-down key 42A is pressed and "Polymer P1" is selected from the pull-down menu 42B, "Polymer P1" is displayed in the input frame 42 (see Figure 7).

The input frame 43 is used to input the compounding ratio of the material entered in the input frame 42. As shown in FIG. 7, when the input frame 43 is selected using the operation unit 24, a numeric keypad input panel 45 is displayed below it. This numeric keypad input panel 45 allows the user to enter the numerical compounding ratio of the material that makes up the predicted rubber. FIG. 7 shows an example in which when "Polymer P1" is entered in the input frame 42, "20 phr" is entered in the input frame 43 as the compounding ratio of that polymer P1. After the compounding ratio is entered in the input frame 43, pressing the registration key 32 registers the compounding ratio of the specific material that makes up the predicted rubber, and the information is displayed in the display frame 44.

By performing the above input operations for each of the materials that make up the predicted rubber, the registration of the compounding ratios of all materials is completed. Figure 8 shows the compounding ratios of each material in predicted rubber Tx, an example of a prediction target, entered and displayed in display frame 44. The compounding ratios of each material in predicted rubber Tx are shown in Table 4.

In Figure 8, when the next key 33 is pressed after the blending ratios for each material belonging to each category have been entered, all registered blending ratios are stored in the memory unit 22, and information on the registered blending ratios is sent to the prediction device 10 by the communication unit 25. In addition, the screen displayed on the display unit 23 changes to the material property input screen 232 (see Figures 9 and 10).

Figures 9 and 10 show the material property input screen 232 (an example of an input screen) displayed on the display unit 23 of the information terminal 20. The material property input screen 232 is an input screen for entering the material property information of the polymer or additive entered on the blend ratio input screen 231.

As shown in FIG. 9, the material property input screen 232 includes an input frame 51 for inputting the material, an input frame 52 for inputting the material properties, an input frame 53 for inputting the characteristic value of the material property (material property value, an example of a feature of the present invention), and a display frame 54 for displaying the input material property value. The material property input screen 232 also has a back key 31, a registration key 32, and a next key 33. The user can input any information into the input frames 51 to 53 by operating the operation unit 24.

The input frame 51 is used to input the polymers or additives that make up the predicted rubber. When the pull-down key 51A in the input frame 51 is pressed using the operation unit 24, a pull-down menu (not shown) containing pre-registered materials (in the case of the predicted rubber Tx, polymers P1, P4, P6, and each additive) is displayed, and by selecting one of these, the target for inputting the material characteristic value is entered into the input frame 51. Figure 9 shows the state in which "Polymer P1" has been entered into the input frame 51.

Input frame 52 is used to input specific material properties whose characteristic values will be entered. Once the input of the material in input frame 51 is complete, input of the material properties possessed by that material becomes possible in input frame 52. For example, if "Polymer P1" is entered in input frame 51, input frame 52 becomes possible to input multiple material properties possessed by Polymer P1. Input frame 52 is provided with pull-down key 52A, and when pull-down key 52A is pressed using operation unit 24, pull-down menu 52B (see Figure 9) containing multiple pre-registered material properties is displayed.

As shown in Figure 9, when one of the multiple material properties displayed in pull-down menu 52B is selected, the selected material property is displayed in input frame 52, and then the material property value can be entered in input frame 53. After the material property value is entered in input frame 53, pressing registration key 32 registers the material property value of the specific material that makes up the predicted rubber, and the information is displayed in display frame 54. Figure 9 shows the state in which the material property values of polymers P1 and P4 contained in predicted rubber Tx shown in Table 5 have been entered and displayed in display frame 54.

When entering the material property values for an additive such as a filler, as shown in Figure 10, select a registered additive in input box 51, select the material property from pull-down menu 52B in input box 52, and then enter the material property value in input box 53. Figure 10 shows the state in which "Additive A2" has been entered in input box 51.

When the next key 33 is pressed after all material property values have been entered on the material property input screen 232, all registered material property values are stored in the memory unit 22, and the registered blending ratio information is sent to the prediction device 10 by the communication unit 25. In addition, the screen displayed on the display unit 23 changes to the output setting screen 233 (see Figure 11).

The first selection processing unit 212 performs processing to select, from the specific performance predicted by the performance prediction unit 111, the predicted performance whose predicted value is to be displayed in the display frame 65. Specifically, the first selection processing unit 212 displays the output setting screen 233 on the display unit 23, and acquires the predicted performance entered into the output setting screen 233 by the operation unit 24. The acquired predicted performance is then set as the specific performance to be predicted by the performance prediction unit 111, and the setting information is sent to the prediction device 10 by the communication unit 25.

Figures 11 to 13 show the output setting screen displayed on the display unit 23 of the information terminal 20. The output setting screen 233 is a setting screen for setting the specific performance to be predicted by the performance prediction unit 111 and the material properties to be displayed in the display frame 65.

As shown in FIG. 11, the output setting screen 233 has an input box 61 for inputting the predicted performance and a display box 62 for displaying the predicted value. The output setting screen 233 also has a back key 31 for returning the displayed content to the previous screen, and an execute key 34 for inputting a prediction execution command. The user can input any information into the input box 61 by operating the operation unit 24.

The input box 61 is used to input the predicted performance to be predicted by the performance prediction unit 111. The input box 61 is provided with a pull-down key 61A, and when the pull-down key 61A is pressed using the operation unit 24, a pull-down menu 61B (see Figure 12) is displayed that includes multiple performance items indicating multiple pre-registered performances.

Figure 12 shows the state when the pull-down key 61A is pressed and the pull-down menu 61B is opened. As shown in Figure 12, the pull-down menu 61B includes multiple performance items selectable as the predicted performance, including hardness, dynamic viscoelasticity at low and high temperatures (complex modulus E*, storage modulus E', loss modulus E", loss tangent tanδ, complex shear modulus G*, storage shear modulus G', loss shear modulus G"), tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300% modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristic value, scorch time, and Mooney viscosity. When one of these performance items is selected, the selected performance item is displayed in the input box 61 and set as the specific performance. Note that Figure 13 shows "low-temperature loss tangent tanδ" selected as the predicted performance and displayed in input box 61.

The second selection processing unit 213 performs a process of selecting the specific material whose material characteristic value (feature amount) is to be displayed in the display frame 65. Specifically, the second selection processing unit 213 acquires the material characteristics of the specific material input to the output setting screen 233 by the operation unit 24. The acquired material characteristics are then set as the display target by the display processing unit 216, and the setting information is sent to the prediction device 10 by the communication unit 25.

As shown in FIG. 13, the output setting screen 233 has an input frame 63 for inputting the specific material and an input frame 64 for inputting the characteristic values of the specific material. The user can input any information into the input frames 63 and 64 by operating the operation unit 24.

The input frame 63 is used to input the specific material to be displayed in the display frame 65, and the input frame 64 is used to input the material properties of the specific material to be displayed in the display frame 65. When the pull-down key 63A in the input frame 63 is pressed, a pull-down menu containing multiple specific materials is displayed, and any specific material can be selected from the menu. When the pull-down key 64A in the input frame 64 is pressed, a pull-down menu containing multiple material properties is displayed, and any material property can be selected from the menu. Figure 13 shows the state in which "Polymer P4" has been selected as the specific material and "Weight average molecular weight (Mw)" has been selected as the material property.

When the predicted performance is selected and the material properties are selected, and then the execute key 34 is pressed, the predicted performance is sent to the prediction device 10 by the communication unit 25 along with an instruction to execute the prediction.

When the display processing unit 216 receives the predicted value transmitted from the prediction device 10, it performs processing to associate the predicted value with the characteristic value of the material properties of a specific material identified from the multiple materials that make up the predicted rubber, and display it on the prediction result display screen 234 (see Figure 14).

In this embodiment, upon receiving the predicted value, the display processing unit 216 displays the predicted value in the display frame 62. Furthermore, as shown in FIG. 14 , the display processing unit 216 associates the predicted value with the characteristic value of the specific material selected by the second selection processing unit 213 and displays the associated value in the display frame 65. The display frame 65 shows a two-dimensional coordinate system with the predicted value on the vertical axis and the characteristic value on the horizontal axis, and coordinate points representing the correspondence between the predicted value and the characteristic value are displayed so that they can be grasped at a glance. In FIG. 14 , when the predicted value of the "loss tangent tan δ" to be predicted is "0.28" and the characteristic value of the material characteristic "weight average molecular weight (Mw)" of the specific material "polymer P4" is "5.0 × 10 ⁵ ," the coordinate point (5.0 × 10 ⁵ , 0.28) is displayed in the two-dimensional coordinate system of the display frame 65.

The display processing unit 216 also displays the material characteristic value of the specific material in the display frame 65. For example, as shown in FIG. 14, a display window 66 capable of displaying the material characteristic value is displayed near the horizontal axis of the two-dimensional coordinate system of the display frame 65, and the numerical value of the material characteristic value, i.e., the coordinate on the horizontal axis, is displayed in the display window 66.

The characteristic value change unit 214 performs processing to change the material characteristic values displayed in the display window 66 of the display frame 65 by the display processing unit 216. The display window 66 is provided with keys 66A and 66B that can increase or decrease the numerical value displayed in the display window 66. As shown in FIG. 15, when the increase key 66A is pressed by the operation unit 24, the characteristic value change unit 214 increases the numerical value according to the time or number of times it is pressed, and when the decrease key 66B is pressed, the numerical value decreases according to the time or number of times it is pressed.

When the material characteristic value displayed in the display window 66 is changed by the characteristic value change unit 214 and the execute key 34 is then pressed, the changed material characteristic value is sent to the prediction device 10 by the communication unit 25 along with an instruction to execute a prediction. In this case, in the prediction device 10, the performance prediction unit 111 re-predicts the physical property value of the specific performance of the rubber being predicted based on material information including the changed material characteristic value, and the prediction device 10 transmits the predicted value to the information terminal 20.

For example, if the value of the weight average molecular weight Mw of the rubber to be predicted, displayed in the display frame 65, is reduced from the initial "5.0×10 ⁵ " to "3.5×10 ⁵ " and then the execute key 34 is pressed (see FIG. 15), the physical property value of the specific performance will again be predicted by the performance prediction unit 111 using the reduced value. Then, as shown in FIG. 16, the predicted value in this case, "0.32", is displayed in the display frame 62, and the coordinate point (3.5×10 ⁵ , 0.32) is displayed in the two-dimensional coordinate system of the display frame 65.

[Prediction processing/display processing]
17, an example of the procedure of the prediction process and the display process executed in the prediction system 100 will be described. The prediction process is executed by the control unit 11 in the prediction device 10, and the display process is executed by the control unit 21 in the information terminal 20. Note that one or more steps included in the prediction process or the display process may be omitted as appropriate, and the steps may be executed in a different order as long as the same operational effect is achieved.

First, in step S11, when the compounding ratio of the rubber to be predicted is input via the compounding ratio input screen 231, the control unit 21 of the information terminal 20 stores the input compounding ratio in RAM or the memory unit 22 as information to be input into the prediction model 123, and then transmits the compounding ratio to the prediction device 10.

In the next step S12, when the physical property values of the material properties of each material that makes up the rubber to be predicted are input via the material property input screen 232, the control unit 21 stores the input physical property values in RAM or the memory unit 22 as information to be input into the prediction model 123, and then transmits the physical property values to the prediction device 10.

After that, the predicted performance is entered into input frame 61 via output setting screen 233, and the material properties are entered into input frame 64 to perform output settings (S13). It is then determined whether a prediction execution command has been entered by pressing execution key 34 (S14).

In step S14, when the prediction execution instruction is input, the control unit 21 transmits the prediction execution instruction along with the prediction performance to the prediction device 10, causing the performance prediction unit 111 of the prediction device 10 to execute the prediction process.

In step S15, the control unit 11 of the prediction device 10 inputs the compounding ratio and the material characteristic values into the input unit of the prediction model 123, and performs a prediction process in which the prediction model 123 predicts the physical property values of the specific performance of the rubber to be predicted. The control unit 11 then transmits the predicted value output from the prediction model 123 by the prediction process to the information terminal 20. Note that step S15 is an example of a prediction step of the present invention.

In step S16, when the control unit 21 of the information terminal 20 receives the predicted value from the prediction device 10, it displays the predicted value in the display frame 62 of the prediction result display screen 234. Furthermore, the control unit 21 displays a two-dimensional coordinate system in the display frame 65, with the predicted value on the vertical axis and the characteristic value of the material characteristic set in step S13 on the horizontal axis, and also displays coordinate points that correspond to the predicted value and the characteristic value (S17). The control unit 21 also displays the numerical value of the characteristic value in the display window 66 of the display frame 65 so that it can be changed (S18). Note that step S17 is an example of a display processing step of the present invention.

Next, in step S19, the control unit 21 determines whether the characteristic value in the display window 66 has changed. For example, if the decrease key 66B is pressed by operating the operation unit 24, it determines that the characteristic value has decreased, and if the increase key 66A is pressed, it determines that the characteristic value has increased.

If the characteristic value is not changed and a prediction end command is input (Yes in S20), the process ends.

On the other hand, if it is determined that the characteristic value has been changed and the execution key 34 has been pressed again (S19 Yes), the control unit 21 sends the changed material characteristic value along with a prediction execution instruction to the prediction device 10, and causes the performance prediction unit 111 to execute the prediction process again (S21). Then, the processing from step S16 onwards is repeated.

As described above, in this embodiment, the display processing unit 216 displays the predicted value of the specific performance of the rubber being predicted, which is the target of prediction, on the display unit 23. As a result, if the user is, for example, a developer of the rubber-like elastomer, the user can easily grasp at a glance the relationship between the characteristic quantities of the specific material that makes up the rubber-like elastomer and the predicted value of the specific performance of the rubber-like elastomer. As a result, the user can use this relationship as a starting point to more quickly identify the composition of a rubber-like elastomer that has the desired performance value, which can greatly contribute to the development of the rubber-like elastomer.

In the above-described embodiment, an example was described in which a two-dimensional coordinate system was displayed in the display frame 65 to display the coordinate points between the predicted value and the characteristic value. For example, as shown in FIG. 18, a prediction graph L1 (an example of a characteristic graph of the present invention) with the characteristic value on the horizontal axis as a variable may be displayed in the two-dimensional coordinate system shown in the display frame 65. In this case, the performance prediction unit 111 may predict a prediction graph L1 corresponding to the possible range of the characteristic value (an example of a quantitative range of the present invention), transmit this to the information terminal 20, and display the prediction graph L1 in the display frame 65. This allows the user to easily grasp at a glance the change trend of the predicted value within the possible range of the characteristic value from the prediction graph L1.

In this case, it is preferable that a change cursor 67 for changing the lower limit of the range and a change cursor 68 for changing the upper limit of the range be provided so that the range of possible characteristic values of the specific material selected by the second selection processing unit 213 can be changed. The change cursors 67 and 68 are moved by the user operating the operation unit 24. When the change cursors 67 and 68 are moved, the range change unit 215 of the control unit 21 changes the range to a range corresponding to the positions of the change cursors 67 and 68 after their displacement. The lower limit of the range changed by the change cursor 67 is displayed in a display window 67A provided on the prediction result display screen 234, and the upper limit of the range is displayed in a display window 67B provided on the prediction result display screen 234. The display processing unit 216 also changes the prediction graph L1 to correspond to the changed range and displays it in the display frame 65. This allows the user to easily grasp the trend in the change of the predicted value in response to changes in the range of possible characteristic values from the changed prediction graph L1 displayed by the display processing unit 216.

In addition, in the above-described embodiment, the coordinate points of the predicted value and the characteristic value corresponding to the target to be predicted are displayed on the horizontal axis of the two-dimensional coordinate system displayed in the display frame 65, but the present invention is not limited to this configuration. For example, when the performance prediction unit 111 performs prediction processing for each of multiple types of target to be predicted, the display processing unit 216 may associate the predicted value corresponding to each target to be predicted with the material characteristic value (feature value) of each of the specific materials corresponding to each predicted value and display them on the display unit. For example, when the performance prediction unit 111 predicts the predicted value for each of two or more target to be predicted, multiple coordinate points corresponding to the multiple target to be predicted may be displayed in the two-dimensional coordinate system. In this case, the user can easily grasp at a glance the relationship between the predicted value and the feature value for each of the different types of target to be predicted.

Also, as shown in FIG. 19, when an arbitrary point on the prediction graph L1 is selected by operating the operation unit 24, the display processing unit 216 may display the predicted value on the vertical axis at that coordinate point in the display frame 62, and the characteristic value on the horizontal axis at that coordinate point in the display window 66.

Furthermore, the coordinates displayed in the display frame 65 are not limited to a two-dimensional coordinate system. For example, as shown in FIG. 20, the Z axis may represent the predicted value, the X axis the weight average molecular weight, which is one of the material properties of the rubber to be predicted, and the Y axis the number average molecular weight, which is one of the material properties of the rubber to be predicted, and the relationship between each numerical value may be displayed in the display frame 65 using a three-dimensional coordinate system.

Furthermore, in the above-described embodiment, an example of a prediction system 100 is shown in which the performance prediction unit 111 is provided in the prediction device 10 and the display processing unit 216 is provided in the information terminal 20, but the present invention is not limited to this configuration. For example, the components provided in the information terminal 20 may be provided in the prediction device 10, or the components provided in the prediction device 10 may be provided in the information terminal 20.

In addition, in the above-described embodiment, the material characteristic values are exemplified as feature quantities displayed in the display frame 65 of the prediction result display screen 234 (see FIG. 14 ), but the feature quantities of the present invention are not limited to this. For example, instead of the material characteristic values, the specific material blend amounts selected by the second selection processing unit 213 may be displayed in the display frame 65 together with the predicted values.

Furthermore, in the above-described embodiment, the polymer composition is unvulcanized rubber, and the rubbery elastomer is a polymer composition that has been vulcanized. However, the present invention is not limited to this configuration. For example, the present invention can predict specific performance of a polymer composition even when the polymer composition is in an unvulcanized state (i.e., unvulcanized rubber).

In addition, in the above-described embodiment, a rubber-like elastomer made of a polymer composition was used as an example of the composition of the present invention. However, the composition is not limited to a rubber-like elastomer, nor is it limited to an unvulcanized polymer composition (unvulcanized rubber). The composition of the present invention may be any organic composition primarily composed of multiple types of organic substances (e.g., a composition having a total organic substance content of more than 10% by mass), and the polymer compositions described above are more preferred. Furthermore, rubber compositions primarily composed of multiple types of rubber materials and resin compositions composed of multiple types of synthetic resins are particularly preferred. In other words, the present invention can also be applied to predicting the physical properties of specific material performance in these compositions.

100: Rubber performance prediction system,
10: Prediction device 11: Control unit 20: Information terminal 21: Control unit 23: Display unit 30: Database 111: Performance prediction unit 121: Control program 123: Prediction model 211: Input reception unit 212: First selection processing unit 213: Second selection processing unit 214: Characteristic value change unit 215: Range change unit 216: Display processing unit 221: Control program

Claims (14)

A physical property prediction system for predicting specific performance of a rubber-like elastomer obtained by vulcanizing a polymer composition composed of a plurality of materials including at least a polymer and an additive, comprising :
a material information storage unit configured to store material information relating to the plurality of materials, the material information including a feature value indicating a predetermined physical property value of a specific material among the plurality of materials;
a prediction unit that predicts a value of the specific performance based on the material information stored in the material information storage unit;
a display processing unit that displays the predicted value predicted by the prediction unit in association with the feature amount stored in the material information storage unit. a feature amount changing unit that changes the feature amount displayed on the predetermined display unit by the display processing unit,
The physical property prediction system according to claim 1 , wherein, when the feature quantity is changed by the feature quantity change unit, the prediction unit re-predicts the value of the specific performance based on the material information including the feature quantity after the change. a first selection unit that selects the specific performance for which the display processing unit displays the predicted value;
The physical property prediction system according to claim 1 , wherein the display processing unit displays the predicted value of the specific performance selected by the first selection unit in association with the feature amount. a second selection unit that selects the specific material for which the feature amount is to be displayed on the display processing unit;
4. The physical property prediction system according to claim 1, wherein the display processing unit displays the predicted value predicted by the prediction unit and the feature amount of the specific material selected by the second selection unit in association with each other. The physical property prediction system described in claim 4, wherein the display processing unit displays, on a predetermined display unit, a characteristic graph showing the predicted values corresponding to the possible quantitative range of the feature of the specific material selected by the second selection unit. a range change unit that changes the possible quantitative range of the feature amount of the specific material selected by the second selection unit,
The physical property prediction system according to claim 5 , wherein the display processing unit causes the characteristic graph corresponding to the quantitative range changed by the range changing unit to be displayed on a predetermined display unit. 7. A physical property prediction system according to claim 1, wherein when values of the specific performance of each of a plurality of types of rubber-like elastomers are predicted by the prediction unit, the display processing unit displays the predicted values corresponding to each rubber-like elastomer in association with the feature quantities of the specific material corresponding to each predicted value. an input receiving unit that receives input of material information of the rubber-like elastic body that is a prediction target by the prediction unit,
8. A physical property prediction system according to claim 1, wherein the prediction unit predicts the value of the specific performance of the rubber- like elastic body based on a prediction model based on training data including feature quantities of a plurality of pieces of material information for each of a plurality of other rubber- like elastic bodies and values of the specific performance for each of the plurality of other rubber-like elastic bodies, and on the material information accepted by the input accepting unit. 9. The physical property prediction system according to claim 1, wherein the specific performance includes at least one of hardness, dynamic viscoelasticity at low and high temperatures, tensile strength, elongation at break, 100% modulus (M100), 200% modulus (M200), 300 % modulus (M300), toluene swelling index, glass transition temperature, abrasion resistance, vulcanization characteristic value, scorch time, and Mooney viscosity. 10. The physical property prediction system according to claim 1, wherein the material information is information about molecules of the polymer and the additive, and includes at least one of weight average molecular weight, number average molecular weight, molecular weight distribution , and degree of branching of molecular chains. 11. The physical property prediction system according to claim 10, wherein the material information includes at least one of the cis-isomer ratio, trans-isomer ratio, amount of oil extension, glass transition temperature, solubility parameter, amount of styrene, amount of vinyl, amount of butadiene rubber, and viscoelastic properties of the polymer . The additives include a filler,
The physical property prediction system according to claim 10 or 11 , wherein the material information includes at least one of a particle size, a CTAB adsorption specific surface area, a BET adsorption specific surface area, and a surface polarity of the filler. A physical property prediction device for predicting specific performance of a rubber-like elastic body obtained by vulcanizing a polymer composition composed of a plurality of materials including at least a polymer and an additive, comprising :
a material information storage unit configured to store material information relating to the plurality of materials, the material information including a feature value indicating a predetermined physical property value of a specific material among the plurality of materials;
a prediction unit that predicts a value of the specific performance based on the material information stored in the material information storage unit;
a display processing unit that displays the predicted value predicted by the prediction unit and the feature amount stored in the material information storage unit in association with each other. A method for predicting a specific performance value of a rubber-like elastomer obtained by vulcanizing a polymer composition composed of a plurality of materials including at least a polymer and an additive, comprising :
a prediction step of predicting a value of the specific performance based on material information about the plurality of materials;
a display processing step of displaying the predicted value predicted in the prediction step in association with a feature value indicating a physical property value of a specified physical property of a specific material among the plurality of materials, the display processing step being executed by one or more processors .