Material information prediction method and material information prediction system

Abstract

To provide a material information prediction method and material information prediction system which can efficiently predict and output multiple pieces of material information through a single prediction process.SOLUTION: A material prediction system provided herein comprises a prediction device and a database. In the material prediction system, a control unit is configured to: acquire physical property information of a prediction target composition with unknown materials and unknown blend ratios of the materials (S31); and predict an identification code and blend information of the prediction target composition on the basis of the acquired physical property information (S32).SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] The present disclosure relates to a material information prediction method and a material information prediction system for predicting material information regarding a material used in manufacturing a composition based on physical property values of the composition.

Background Art

[0002] Conventionally, prediction methods for obtaining specific performances (physical properties) of rubber compositions, which are materials such as tires mounted on automobiles, are known. Further, using the names of a plurality of raw materials used in manufacturing a rubber composition (for example, product names or model numbers specific to the raw materials) and their blending ratios as explanatory variables, and using the physical property data of the rubber composition as an objective function, a physical property data prediction method is known in which a computer is machine-learned to construct a prediction model and the physical property data is predicted using the prediction model.

[0003] Further, a method is disclosed in which, using a neural network technique, the relationship between a factor group and a characteristic group of experimental data such as design and blending is learned, characteristics are estimated from arbitrary factor data, and factors are estimated from arbitrary characteristic data (see Patent Document 1).

[0004] Further, as another physical property data prediction method, a technique is disclosed in which, without using a product name or model number specific to a raw material as identification information of the raw material, an identification name obtained by classifying a plurality of raw materials having similar material characteristics into one is used (see Patent Document 2).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] Incidentally, even if prediction results of the materials constituting the rubber composition and their blending ratios are obtained based on the physical property values of the rubber composition, the materials are not necessarily the most suitable materials for manufacturing the rubber composition. For example, depending on conditions such as cost, difficulty in obtaining, and throughput, inappropriate materials may be predicted. On the other hand, if it is possible to predict not only the predicted materials but also materials that do not change the physical property values of the rubber composition even when substituting the materials, or materials whose variation range remains within the allowable range even if they change, developers can easily find the most suitable materials in designing the rubber composition and contribute to the development of the rubber composition. However, in order to obtain a plurality of prediction results, the conventional prediction method requires performing a plurality of prediction processes, and the prediction process takes an enormous amount of time.

[0007] An object of the present disclosure is to provide a material information prediction method and a material information prediction system capable of efficiently predicting and outputting a plurality of material information by a single prediction process.

Means for Solving the Problems

[0008] A material information prediction method according to an aspect of the present disclosure is a material information prediction method for predicting material information regarding a material used in manufacturing the composition based on physical property values of the composition, including: a search step of searching for a plurality of similar material identification information corresponding to a plurality of the physical property information that are similar to each other by a predetermined optimization process with reference to a first data set including the physical property information including the physical property values of a plurality of compositions, material identification information indicating the materials used in manufacturing the composition, and blending information regarding blending of a plurality of materials; an assignment step of assigning an identifier common to the plurality of searched similar material identification information; and a prediction step of inputting the physical property values of a prediction target composed of the composition into a material prediction model learned using a second data set for learning in which the similar material identification information in the first data set is replaced with the identifier, and outputting the identifier corresponding to the material of the prediction target from the material prediction model and the blending information of the material to which the identifier is assigned.

Advantages of the Invention

[0009] According to the present disclosure, it is possible to efficiently predict and output a plurality of material information by a single prediction process.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. Note that the following embodiments are an example of embodying the present invention and do not limit the technical scope of the present invention.

[0012] FIG. 1 is a diagram showing the configuration of a material prediction system 100 according to an embodiment of the present disclosure. The material prediction system 100 of the present embodiment mainly includes a prediction device 10 and a database 30. Note that the material prediction system 100 is an example of the material information prediction system of the present disclosure and is also a system for realizing the material information prediction method of the present disclosure.

[0013] The material prediction system 100 is configured to be able to predict material information regarding materials used in the production of a polymer composition (an example of the composition of the present disclosure) composed of a plurality of materials including polymers and additives such as polymers and additives. The material information includes the materials required for the production of the polymer composition and the blending ratio of the materials. In the present embodiment, the material prediction system 100 is configured to predict the materials constituting the polymer composition and the blending ratio of the materials using the property information indicating the physical properties and their physical property values for the polymer composition whose physical properties and their physical property values are known. Note that the physical property is a property of the polymer composition, and the physical property value is a physical quantity obtained by quantifying the degree of the property.

[0014] Hereinafter, as a polymer composition to be predicted by the material prediction system 100, a rubber composition (rubbery elastic body) obtained by vulcanizing a compounded rubber (kneaded material) obtained by kneading a plurality of materials including a polymer such as raw rubber and an additive such as silica will be exemplified, and various processes executed by the material prediction system 100 will be described. Needless to say, the rubber composition is merely an example of the composition to be predicted, and the material prediction system 100 may be configured to predict each material constituting the composition (an example of the composition of the present disclosure) as a rubbery elastic body formed without going through a vulcanization process and the blending ratio of those materials.

[0015] Here, as described above, the rubber composition is obtained by vulcanizing the compounded rubber, and specifically, it is a rubber material used in the manufacture of tire products such as pneumatic tires mounted on vehicles such as automobiles. That is, the material prediction system 100 of the present embodiment can predict each material constituting the rubber composition of the tire product and the blending ratio of each material.

[0016] The prediction result by the material prediction system 100 is used, for example, in the development of the rubber composition and the development of the tire product manufactured using this rubber composition. If the prediction result of the material information of the rubber composition is obtained from the physical property information of the rubber composition, the developers of the rubber composition and the tire product can grasp the general material information of the test products without performing material analysis processing on the test products of the rubber composition and the tire product. Further, if the material information of the rubber composition and the tire product is obtained, the developers can obtain the blending information such as the types of materials constituting the rubber composition and the tire product and their blending ratios, so that it is possible to manufacture a rubber composition or a tire product having the same physical property values as the rubber composition and the tire product to be predicted.

[0017] In the present embodiment, a rubber material used in the manufacture of the tire product is exemplified as an example of the rubber composition. However, for example, the rubber composition may be the tire product itself. In this case, the material prediction system 100 predicts the material information of the tire product based on the tire performance (an example of physical properties) and its performance value (an example of physical property values) of the tire product. Further, the rubber composition may be the industrial rubber product itself such as a vibration-proof rubber or a rubber material used in the manufacture of industrial rubber products.

[0018] As described above, the compounded rubber is an unvulcanized rubber obtained by kneading a plurality of materials including one or more polymers and one or more additives. The rubber composition is obtained by vulcanizing the compounded rubber.

[0019] The polymer is, for example, an unvulcanized raw rubber compounded into the compounded rubber. Examples of the raw rubber include natural rubber (NR), isoprene rubber (IR), butadiene rubber (BR), styrene-butadiene rubber (SBR), and the like.

[0020] Examples of the additives include fillers (filler agents) such as carbon black and silica, coupling agents, anti-aging agents, vulcanization accelerators, oils, zinc oxide, stearic acid, sulfur, processing aids, and the like.

[0021] [Configuration of Material Prediction System 100] As shown in FIG. 1, the material prediction system 100 includes a prediction device 10 and a database 30, and these are connected by a network N1 so as to be capable of data communication. 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.

[0022] In this embodiment, a configuration in which the database 30 is connected to the network N1 is illustrated. However, for example, the database 30 may be provided in the prediction device 10. That is, a storage device that substitutes for the database 30 may be provided in the prediction device 10. In this case, the material information prediction system of the present disclosure is realized by the prediction device 10 alone.

[0023] The prediction device 10 is an element constituting the material prediction system 100. The prediction device 10 predicts the material information regarding the material used in the production of the rubber composition to be predicted by using various information such as input information input to the prediction device 10 and a prediction model 123 (see FIG. 4) constructed in advance. Further, the prediction device 10 outputs the prediction result to the information terminal 20 connected to the network N1.

[0024] In addition, the prediction device 10 generates a material classification list 32 (see FIG. 3) described later using various information such as input information input to the prediction device 10 and an optimization model 124 (see FIG. 4) described later. Further, a learning dataset 33 (see FIG. 6) is generated as teaching data necessary for machine learning of the prediction model 123. Note that the learning dataset 33 is an example of the second dataset of the present disclosure.

[0025] The prediction device 10 is an information processing device capable of executing various arithmetic processes. For example, it is a server computer, a cloud server, or a personal computer connected to the network N1. Note that the prediction device 10 is not limited to a single computer, and may be a computer system in which a plurality of computers cooperate to operate, or a cloud computing system. Further, various processes executed by the prediction device 10 may be distributed and executed by one or a plurality of processors. A program or computer software for operating the material prediction system 100 is installed in the prediction device 10.

[0026] The information terminal 20 is an information processing device or terminal device used by a user. The information terminal 20 is a so-called desktop personal computer, a notebook personal computer, or a portable terminal such as a smartphone or a tablet terminal that can be carried around. The user operates the information terminal 20 to input various information necessary for various processes executed in the prediction device 10 to the prediction device 10. Further, the information terminal 20 displays the prediction result and the like transmitted from the prediction device 10 on the display screen. Therefore, a program or computer software for transmitting the various information to the prediction device 10 in cooperation with the material prediction system 100 and displaying the prediction result on the display screen is installed in the information terminal 20.

[0027] The database 30 is a collection of various data handled in the material prediction system 100 stored in a storage medium based on a predetermined data management method. The database 30 is managed in various forms such as a storage device, an information processing device, a cloud server, a data server, etc. that are connected to the network N1 so as to enable data communication. The database 30 stores a reference data set 31 (an example of the first data set of the present disclosure) and a material classification list 32.

[0028] The reference data set 31 is a data set used for constructing (generating) a prediction model 123 (see FIG. 4) used in the prediction process by the prediction device 10. In other words, the reference data set 31 is a data set used to generate a learning data set 33 (see FIG. 6) used for machine learning of the prediction model 123 described later.

[0029] Specifically, the reference data set 31 includes material identification information, compounding information, and physical property information of each of a large number of the rubber compositions whose physical properties and physical property values are known. Hereinafter, the rubber composition whose physical properties and physical property values are known may be referred to as a sample rubber.

[0030] The sample rubber is, for example, a rubber product (such as a tire product or an industrial rubber product) manufactured as a product so far, a prototype rubber product manufactured during research and experiments for the development of the rubber product, or a test piece made of the rubber composition manufactured during the research.

[0031] The reference data set 31 is an aggregate of data in which a large number of composition characteristic data 31A corresponding to various sample rubbers are collected. Each composition characteristic data 31A includes the material identification information, the compounding information, and the physical property information regarding the corresponding sample rubber. This reference data set 31 is stored in the database 30.

[0032] Here, the material identification information is information for identifying a plurality of materials used in the production of the sample rubber, and examples thereof include a material name for specifying a material, a product name, a product model number, a serial number, and the like.

[0033] The compounding information is information indicating the compounding ratio of a plurality of materials used in the production of the sample rubber.

[0034] The physical property information is information indicating the physical properties and physical property values of the sample rubber.

[0035] FIG. 2 is a diagram illustrating one composition characteristic data 31A included in the reference dataset 31. As shown in FIG. 2, the composition characteristic data 31A includes a material name and its compounding ratio for each material, and also includes a physical property name and a physical property value as the physical property information of the sample rubber.

[0036] In the composition characteristic data 31A of FIG. 2, "TSR20", "SBR1502", and "BR10B" are material names for identifying polymers (raw rubbers), "N220" is a material name for identifying carbon black as a filler (filler / reinforcing agent), "VN3" is a material name for identifying silica as a filler (filler / reinforcing agent), and "Si69" is a material name for identifying a coupling agent as an additive. Also, as material names for identifying other additives, "5% oil sulfur", "zinc oxide", "stearic acid", "vulcanization accelerator", etc. are shown in the composition characteristic data 31A.

[0037] Further, in the composition characteristic data 31A, numerical values representing the compounding ratio of each material in parts by mass are shown as the compounding information of each material. Specifically, the compounding ratio of each material indicates the ratio of the compounding amount (parts by mass) of each material when the total mass parts of each polymer contained in the sample rubber is set to 100. The unit used for the compounding ratio is represented by phr (: per hundred rubber).

[0038] In addition, in the composition characteristic data 31A, as various physical properties of the sample tire, for example, Mooney viscosity, scorch time measured by a Mooney viscometer, vulcanization characteristic values (induction time tC(10), 50% vulcanization time tC(50), 90% vulcanization time tC(90), etc.) measured by a vulcanization tester (curastometer (registered trademark)), hardness, tensile strength, glass transition temperature, high-temperature loss tangent tanδ, low-temperature loss tangent tanδ, elastic modulus (storage elastic modulus E′, loss elastic modulus E″, storage shear elastic modulus G′, loss shear elastic modulus G″, etc.), fracture strength, and physical property names and their physical property values of wear performance are shown.

[0039] Note that the composition characteristic data 31A included in the reference data set 31 is not limited to that shown in FIG. 2. For example, kneading conditions when kneading a plurality of materials of the sample rubber, vulcanization conditions for vulcanizing the compounded rubber after kneading, etc. may be included in the composition characteristic data 31A. The kneading conditions are, for example, the volume of the chamber of the kneader, the filling amount of the polymer composition during kneading, the kneading time, the kneading temperature, etc. Also, the vulcanization conditions are, for example, the vulcanization time, vulcanization temperature, etc. set for the compounded rubber before vulcanization of each sample rubber. Further, the composition characteristic data 31A may include all identification information (for example, product name, product model number, serial number, manufacturer information, etc.) used in the market as identification information of each material, and function information indicating characteristics such as the role and function of each material.

[0040] The material classification list 32 (see FIG. 1) is data (material classification data) referred to when generating a learning data set 33 (see FIG. 6) used for machine learning of the prediction model 123 described later. The material classification list 32 is stored in the database 30 together with the reference data set 31. Note that the material classification list 32 may be stored in a storage device different from the reference data set 31.

[0041] FIG. 3 is a diagram showing an example of the material classification list 32. As shown in FIG. 3, the material classification list 32 is table data (material classification data) in which material types divided by each material type for each material category, material names (material identification information) belonging to the material types, and identification codes (an example of the identifier of the present disclosure) assigned to the material names are represented. In the material classification list 32 of FIG. 3, two categories of polymer and filler roughly classified based on their uses are defined as the material categories. Further, the material category "polymer" is classified into four material types (NR, IR, SBR, BR), and the material category "filler" is classified into two material types (carbon, silica). In FIG. 3, actual names are shown in part in the column of the material name, and fictional names are shown in part.

[0042] In the present embodiment, the material classification list 32 includes all the material names included in the reference data set 31. The material classification list 32 is generated by the control unit 11 (see FIG. 4) of the prediction device 10 as will be described later. For example, the control unit 11 extracts all the material names from the reference data set 31, specifies the material type to which each material belongs based on each material name, and further specifies the material category to which each material type belongs based on the type name of each material type. Then, table data (material classification list 32) in which the material category, the material type, and the material name are summarized in a list format is created.

[0043] On the other hand, the identification code included in the material classification list 32 is issued when the prediction model construction process (see FIGS. 7 and 8) described later is executed and is assigned to each material name. The method of assigning the identification code will be described later.

[0044] Note that the material classification list 32 is not limited to that shown in FIG. 3. For example, not only two categories of polymer and filler, but also material types and material categories corresponding to all materials that can be selected as materials (polymers, additives, etc.) of the rubber composition may be defined.

[0045] [Prediction device 10] Hereinafter, with reference to FIG. 4, the configuration of the prediction device 10 will be described. FIG. 4 is a block diagram showing the configuration of the prediction device 10.

[0046] The prediction device 10 is for realizing the material prediction system 100 of the present embodiment, and uses the prediction model 123 described later to perform a process of predicting the materials and the blending ratios of the materials used in the production of a rubber composition (hereinafter sometimes referred to as a prediction target composition) whose materials and blending ratios of the materials are unknown.

[0047] Further, the prediction device 10 generates a learning data set 33 (see FIG. 6) necessary for machine learning of the prediction model 123, and performs a process of constructing (generating) the prediction model 123 using the learning data.

[0048] As shown in FIG. 4, the prediction device 10 includes a control unit 11, a storage unit 12, a communication unit 13, a display unit 14, and an operation unit 15.

[0049] The communication unit 13 is a communication interface for connecting the prediction device 10 to the network N1 and performing data communication with each device connected to the network N1 according to a predetermined communication protocol. Specifically, the communication unit 13 performs data communication with the information terminal 20 and the database 30 through the network N1.

[0050] The display unit 14 is a display device such as a liquid crystal display or an organic EL display that displays various types of information.

[0051] The operation unit 15 is an input device such as a mouse, a keyboard, or a touch panel that receives an operation of an operator.

[0052] The storage unit 12 is a non-volatile storage medium such as an HDD, an SSD, or a flash memory that stores various types of information and data. The storage unit 12 stores a control program 121, a prediction model 123, and an optimization model 124. Note that the prediction model 123 and the optimization model 124 may each be realized as an electronic circuit including a memory in which they are stored.

[0053] The control program 121 is non-temporarily recorded on a computer-readable recording medium such as a CD or a DVD, and may be read from the recording medium by a reading device (not shown) such as a CD drive or a DVD drive that is electrically connected to the prediction device 10 and then copied to the storage unit 12. Alternatively, the control program 121 may be read from an external storage connected to the network N1, input through the communication unit 13, and then copied to the storage unit 12.

[0054] The control program 121 is a program for executing prediction processing using the prediction model 123 or for executing optimization processing (see FIGS. 7 and 8) described later using the optimization model 124. Note that instead of the control program 121, a program for causing the control unit 11 to execute the prediction processing and a program for causing the control unit 11 to execute the optimization processing may be separately stored in the storage unit 12.

[0055] The prediction model 123 is a learned model used for prediction processing for predicting materials used in manufacturing the composition to be predicted and the blending ratio of each material (material information), and derives predicted values of the materials and the blending ratio (material information) of the rubber composition (composition to be predicted) that is the prediction target. In the present embodiment, when the property information of the composition to be predicted that is the prediction target is input to the input unit of the prediction model 123 as an explanatory variable, the prediction model 123 predicts the materials and the blending ratio that constitute the composition to be predicted, and outputs the predicted values from the output unit of the prediction model 123.

[0056] In this embodiment, the prediction model 123 outputs the identification code corresponding to the material used in the production of the composition to be predicted and the blending information indicating the blending ratio of similar materials to which the identification code is assigned. Note that the prediction model 123 may include a well-known function (prediction function) that returns a predicted value (output data) for an input value (input data).

[0057] Here, the similar materials are materials whose characteristics of the material itself (physical property values of the material) and influences in the polymer composition are approximated to each other. That is, each similar material belongs to the same material group, and each similar material is a material whose characteristics of the material itself (physical property values) and influences in the polymer composition are approximated to each other. In other words, even if the material used in the production of the polymer composition is replaced with another similar material within the same material group, the physical property values of the polymer composition produced with the replaced similar material do not change significantly from those of the polymer composition produced with the material before replacement, and the difference is within an acceptable range.

[0058] Hereinafter, a group in which a plurality of the similar materials are classified together is defined as the material group, and an identifier for identifying the material group is defined as the identification code.

[0059] The prediction model 123 is constructed (generated) by a prediction model construction unit 200 described later. In this embodiment, the prediction model 123 is constructed (generated) by machine learning by the control unit 11 based on the learning data set 33 (see FIG. 6) generated by the learning data set generation unit 204 (see FIG. 5) and a predetermined algorithm. Further, when the learning data set 33 is updated, the control unit 11 re-learns and updates the prediction model 123 each time. The details of the construction method of the prediction model 123 will be described later.

[0060] As algorithms necessary for constructing the prediction model 123, for example, algorithms such as multiple regression, generalized linear regression, principal component regression, ridge regression, lasso regression, kernel regression, random forest regression, Gaussian process regression, multi-layer neural network, clustering, support vector machine, and RBF network defined by radial basis functions are preferably used. Further, the prediction model 123 may be constructed by deep learning (deep learning) that multiplies the intermediate layer of the neural network. Note that the prediction model 123 may use one of the above-described algorithms, or may use a plurality of algorithms. In the present embodiment, the generalized linear regression, which is excellent in the balance between accuracy and cost, is used as the algorithm of the prediction model 123.

[0061] The optimization model 124 is used for the optimization process (see FIGS. 7 and 8) by the optimization processing unit 202 (see FIG. 5) described later. In the present embodiment, the optimization model 124 refers to the reference data set 31 including the physical property information, the material identification information, and the compounding information, and searches for a plurality of pieces of material identification information (hereinafter referred to as similar material identification information) indicating a plurality of similar materials in which the characteristics (physical property values of the materials) of the materials themselves are in an approximate relationship with each other. The details of the optimization process by the optimization processing unit 202 (see FIG. 5) will be described later.

[0062] The optimization model 124 preferably uses algorithms such as a genetic algorithm (GA), list search, particle swarm optimization (PSO), response surface methodology (RSM), simulated annealing (SA), multi-objective optimization algorithm, and Bayesian optimization. In this embodiment, the optimization model 124 may use one of the above algorithms or may use a plurality of algorithms. In this embodiment, from the viewpoint of computational efficiency, the Bayesian optimization is used as the algorithm of the optimization model 124.

[0063] The control unit 11 controls the operations of each part of the prediction device 10. The control unit 11 includes control devices such as a CPU, a ROM, and a RAM. The CPU is a processor that executes various arithmetic processes. The ROM is a non-volatile storage medium in which control programs such as BIOS and OS for causing the CPU to execute various arithmetic processes are stored in advance. The RAM is a volatile or non-volatile storage medium that stores various information and is used as a temporary storage memory (working area) for various arithmetic processes executed by the CPU. The control unit 11 controls the prediction device 10 by causing the CPU to execute various control programs stored in advance in the ROM or the storage unit 12.

[0064] As shown in FIG. 4, the control unit 11 includes various processing units such as a data acquisition processing unit 111 (an example of the data acquisition unit of the present disclosure), a prediction processing unit 114 (an example of the prediction processing unit of the present disclosure), a similar material extraction unit 115, an output processing unit 116, and a prediction model construction unit 200.

[0065] The control unit 11 functions as the various processing units by the CPU executing various arithmetic processes according to the control program. The control unit 11 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 11 may be configured by electronic circuits. Also, the control program may be a program for causing a plurality of processors to function as the various processing units.

[0066] The data acquisition processing unit 111 performs a process of acquiring the physical property information including the physical properties of the prediction target composition whose material and blending ratio of the material are unknown and the physical property values of the physical properties.

[0067] The physical property information of the prediction target composition is input by the user in the information terminal 20 when obtaining a predicted value of the material identification information indicating the material used for manufacturing the prediction target composition and the blending information indicating the blending ratio regarding the blending of a plurality of materials required for manufacturing the prediction target composition. For example, when the physical property information input by the user in the information terminal 20 is transferred from the information terminal 20 to the prediction device 10 through the network N1, the data acquisition processing unit 111 acquires the physical property information.

[0068] Note that the data acquisition processing unit 111 may acquire the physical property information of the prediction target composition, for example, by accessing an external storage device indicated by the address information input from the information terminal 20, reading out the physical property information of the prediction target composition stored in the external storage device, and storing it in the storage unit 12.

[0069] The prediction processing unit 114 performs a process of predicting the identification code corresponding to the material of the prediction target composition and the blending information of the similar material to which the identification code is assigned based on the physical property information acquired by the data acquisition processing unit 111.

[0070] Specifically, the prediction processing unit 114 inputs the physical property information acquired by the data acquisition processing unit 111 into the input unit of the prediction model 123, and causes the prediction model 123 to predict the identification code corresponding to the material used in the production of the composition to be predicted, and the compounding information of the similar material to which the identification code is assigned. The identification code and the compounding information of the composition to be predicted are output as a prediction result from the output unit of the prediction model 123.

[0071] The physical properties input to the prediction processing unit 114 are, for example, Mooney viscosity, scorch time measured by a Mooney viscometer, vulcanization characteristic values (induction time tC(10), 50% vulcanization time tC(50), 90% vulcanization time tC(90), etc.) measured by a vulcanization tester (curastometer (registered trademark)), hardness, tensile strength, glass transition temperature, high-temperature loss tangent tanδ, low-temperature loss tangent tanδ, elastic modulus (storage elastic modulus E′, loss elastic modulus E″, storage shear elastic modulus G′, loss shear elastic modulus G″, etc.), fracture strength, and wear performance, which are any plurality or all of them.

[0072] The similar material extraction unit 115 performs a process of extracting a plurality of pieces of the similar material identification information to which the same identification code as the identification code predicted by the prediction processing unit 114 is assigned from the material classification list 32 in which the similar material identification information indicating the similar material is associated with the identification code.

[0073] The similar material identification information extracted by the similar material extraction unit 115 is output to the information terminal 20 or other external devices via the communication unit 13 by the output processing unit 116 for display on the display unit of the information terminal 20 or other external devices.

[0074] The prediction model construction unit 200 executes a process of constructing (generating) a learned prediction model 123 by learning the prediction model 123 by machine learning based on a pre-generated learning data set 33 (see FIG. 6) and a predetermined algorithm.

[0075] FIG. 5 is a block diagram showing the configuration of the prediction model construction unit 200. As shown in FIG. 5, the prediction model construction unit 200 includes various processing units such as a condition setting unit 201, an optimization processing unit 202, an identification code assignment unit 203, a learning data set generation unit 204, and an evaluation determination unit 205.

[0076] The condition setting unit 201 sets setting conditions such as constraint conditions for the target value of optimization by the optimization processing unit 202 and inputs them to the optimization processing unit 202. The constraint condition is, for example, a constraint condition for determining the common identification code assigned to each material, and is, for example, the maximum value of the number of types of materials belonging to a specific material category. For example, when the number of types of SBR is 42, the possible values of the identification code are at most 42 and at least 1. In this case, the condition setting unit 201 sets the upper limit value of the maximum number of searches for the plurality of similar material identification information, in other words, the possible number of the common identification code, to 42 as the constraint condition. Note that the constraint condition may be a condition for restricting the blending ratio of each material within a predetermined range, a condition for restricting the physical property value of a specific physical property within a predetermined range, or the like.

[0077] The optimization processing unit 202 refers to the reference data set 31 in the database 30 and performs a process of searching for a plurality of the similar material identification information corresponding to a plurality of physical property information in which the characteristics of the material itself (physical property values of the material) are approximated to each other in each of the plurality of composition characteristic data 31A.

[0078] In the present embodiment, when an execution instruction for executing the optimization process is input, the optimization processing unit 202 enables the optimization model 124 and inputs the constraint condition and the reference data set 31 to the optimization model 124 to execute a search process for an optimal solution by the optimization model 124. By repeating this search process, the design variables for which the calculated value of the above objective function becomes the minimum value are specified. That is, the similar material identification information for achieving the target value is specified.

[0079] The identification code assignment unit 203 performs an assignment process of assigning the identification code common to the plurality of searched similar material identification information. For example, when the optimization processing unit 202 obtains a solution that the physical property values of the materials indicated by the material names "NR1101" and "NR1102" are approximated to each other, in the material classification list 32, a common identification code "NR01" is assigned as the identification code corresponding to these material names and registered in the material classification list 32.

[0080] The learning data set generation unit 204 performs a data generation process of generating a learning data set 33 for training the prediction model 123 by replacing the similar material identification information in the reference data set 31 with the identification code assigned by the identification code assignment unit 203. FIG. 6 is a diagram showing an example of the learning data set 33, and the material names of the polymer and the filler in each composition characteristic data 31A of the reference data set 31 shown in FIG. 2 are replaced with the identification code. That is, the learning data set 33 includes a plurality of composition characteristic data 33A in which the material names are converted into the identification code. FIG. 6 shows a plurality of composition characteristic data 33A in which the material names are converted into the identification code.

[0081] The learning data set 33 generated by the learning data set generation unit 204 is used as teacher data for machine learning of the prediction model 123. That is, when it is necessary for the control unit 11 to machine-learn the prediction model 123, the learning data set 33 is input to the prediction model 123 to machine-learn the prediction model 123.

[0082] The evaluation determination unit 205 inputs the physical property information included in the verification data prepared in advance into the learned prediction model 123, and based on the blending ratio included in the blending information among the prediction results output from the prediction model 123, performs a process of evaluating the prediction result by the prediction model 123. Here, the verification data is data including the physical property information of the verification rubber composition prepared for verification, the identification code corresponding to the verification rubber composition, and the blending information.

[0083] An example of the prediction evaluation value indicating the evaluation of the prediction result is, for example, the correlation coefficient (:R) between the blending ratio obtained by inputting the physical property information included in the verification data into the prediction model 123 and the known blending ratio (measured value) included in the verification data, and the root mean square error (:RMSE) between the predicted value and the known blending ratio (measured value). The evaluation determination unit 205 calculates the correlation coefficient and the root mean square error between the predicted value and the known allocation ratio.

[0084] Here, the correlation coefficient is an index indicating the linear relationship between the predicted value and the known blending ratio, and takes a value from -1 to +1. The larger the correlation coefficient, the higher the prediction accuracy by the prediction model 123 is evaluated. Let the correlation coefficient be R, then the correlation function is calculated by the following calculation formula.

[0085]

Equation

[0086] Also, the root mean square error is an index indicating the magnitude of the error of the predicted value, and is expressed as the square root of the average value of the squares of the differences between the predicted value and the known blending ratio. The smaller the root mean square error, the higher the prediction accuracy by the prediction model 123 is evaluated. Let the root mean square error be RMSE, then the root mean square error is calculated by the following calculation formula.

[0087]

Equation

[0088] In this embodiment, based on the predicted evaluation value, the search process for the optimal solution by the optimization model 124, the subsequent allocation process by the identification code allocation unit 203, the data generation process by the learning data set generation unit 204, and the learning process of the prediction model 123 using the learning data set 33 are repeatedly executed.

[0089] Specifically, since the larger the correlation coefficient, the higher the prediction accuracy by the prediction model 123, the control unit 11 repeatedly executes the search process, the allocation process, the data generation process, and the learning process until the calculated value of the correlation coefficient becomes equal to or greater than a predetermined first set value. In this case, the prediction model 123 is learned using the learning data set 33 when the calculated value of the correlation coefficient becomes equal to or greater than the first set value, and the construction of the learned prediction model 123 is completed.

[0090] Alternatively, the control unit 11 may repeatedly execute the search process, the allocation process, the data generation process, and the learning process until the number of times the maximum value of the calculated value of the correlation coefficient has not been updated reaches a predetermined first set number of times. For example, even if the calculation of the first set number of times is performed after the previous update, if a calculation result larger than the calculated value at the previous update cannot be obtained, it is determined that the calculated value of the correlation coefficient has converged. In this case, the prediction model 123 is learned using the learning data set 33 at the time of the previous update, and the construction of the learned prediction model 123 is completed.

[0091] Further, since the smaller the root mean square error, the higher the prediction accuracy by the prediction model 123, the control unit 11 repeatedly executes the search process, the allocation process, the data generation process, and the learning process until the calculated value of the root mean square error becomes less than a predetermined second set value. In this case, the prediction model 123 is learned using the learning data set 33 when the calculated value of the root mean square error becomes less than the second set value, and the construction of the learned prediction model 123 is completed.

[0092] Alternatively, the control unit 11 may repeatedly execute the search process, the allocation process, the data generation process, and the learning process until the number of times the minimum value of the calculated root mean square error has not been updated reaches a second set number of times determined in advance. For example, even if the second set number of times is calculated after the previous update, if a calculation result smaller than the calculated value at the previous update cannot be obtained, it is determined that the calculated value of the root mean square error has converged. In this case, the prediction model 123 is learned using the learning data set 33 at the previous update, and the construction of the learned prediction model 123 is completed.

[0093] Regarding the correlation coefficient, a range equal to or greater than the first set value is an example of the allowable range of the present disclosure. Also, regarding the root mean square error, a range less than the second set value is an example of the allowable range of the present disclosure. The first set value and the second set value can be arbitrarily set according to the prediction accuracy required for the prediction model 123.

[0094] In the present embodiment, the control unit 11 repeatedly executes the search process, the allocation process, the data generation process, and the learning process until the correlation coefficient is equal to or greater than the first set value and the root mean square error is less than the second set value.

[0095] Note that the control unit 11 may repeatedly execute the search process, the allocation process, the data generation process, and the learning process until the number of times the maximum value of the calculated correlation coefficient has not been updated reaches the first set number of times and the number of times the minimum value of the calculated root mean square error has not been updated reaches the second set number of times.

[0096] Also, in the present embodiment, even if the predicted evaluation value is outside the allowable range, when the number of executions of the search process reaches a preset upper limit number of times, the control unit 11 may terminate the construction of the prediction model 123 without executing the search process and subsequent processes any further. For example, even if the correlation coefficient is less than the first set value or the root mean square error is greater than or equal to the second set value, when the number of executions of the search process reaches the set number of times, the prediction model 123 is learned using the learning data set 33 at that time, and the construction of the prediction model 123 is completed.

[0097] Note that the predicted evaluation value may be, for example, a difference value between the predicted value and the blending ratio included in the known blending information. In this case, the search process and subsequent processes are executed until the difference value becomes less than a preset threshold value.

[0098] [Prediction Model Construction Process] Hereinafter, with reference to the flowchart of FIG. 7, an example of the procedure of the prediction model construction process executed in the material prediction system 100 will be described together with the material information prediction model construction method of the present disclosure. Note that one or more steps included in the prediction model construction process described below may be appropriately omitted, and the execution order of each step may be different within a range that produces the same operational effects. Also, in the following description, the execution subject of each step is described as the control unit 11, but the execution subject of each step may be the prediction model construction unit 200 or one or more processors included in the prediction model construction unit 200.

[0099] When an execution instruction for executing the optimization process is input, the control unit 11 inputs the reference data set 31 read from the database 30 and the constraint condition set by the condition setting unit 201 into the optimization model 124 of the optimization processing unit 202 (S11).

[0100] Then, in the next step S12, the control unit 11 enables the optimization model 124 and executes a search process for an optimal solution by the optimization model 124 based on the reference data set 31 and the constraint conditions. The optimization model 124 performs a process of searching for a plurality of pieces of the similar material identification information corresponding to a plurality of pieces of physical property information in which the properties of the material itself (physical property values of the material) are approximated to each other in the composition property data 31A of the reference data set 31 based on the input reference data set 31 and the constraint conditions. Note that step S12 is an example of the search step of the present disclosure.

[0101] When a solution is output by the search process by the optimization model 124 (S13), the control unit 11 assigns a common identification code to the plurality of pieces of the similar material identification information that have been searched (S14). Note that step S14 is an example of the assignment step of the present disclosure.

[0102] In the next step S15, the control unit 11 replaces the similar material identification information in the reference data set 31 with the common identification code to generate a learning data set 33 for learning the prediction model 123. Note that step S15 is an example of the learning data set generation step.

[0103] In step S16, the control unit 11 inputs the generated learning data set 33 as teacher data to the prediction model 123 to learn the prediction model 123. Note that step S16 is an example of the learning step.

[0104] Thereafter, the learned prediction model 123 is stored in the storage unit 12 to enable prediction processing by the prediction model 123 (S17). Thereby, a series of prediction model construction processes are completed.

[0105] [Other processing examples of prediction model construction processing] Hereinafter, with reference to the flowchart of FIG. 8, other processing examples of the prediction model construction processing will be described. Note that for the same processing procedures as those shown in the processing procedure of FIG. 7, the same step numbers are assigned in FIG. 8 to omit the detailed description thereof.

[0106] As shown in FIG. 8, when each process shown in steps S11 to S16 is sequentially performed, then, in step S201, the control unit inputs, as explanatory variables, the identification code and the physical property information corresponding to the rubber composition with known compounding information to the input unit of the learned prediction model 123 learned in step S16, and causes the prediction model 123 to perform a prediction process. Thereafter, in step S202, the control unit 11 evaluates the prediction result by the prediction model 123 based on the predicted value output from the prediction model 123.

[0107] In the next step S203, the control unit 11 determines whether or not the prediction evaluation value is within a predetermined allowable range. In the present embodiment, the control unit 11 determines whether or not the correlation coefficient is equal to or greater than the first set value and the root mean square error is less than the second set value. That is, when the correlation coefficient is equal to or greater than the first set value and the root mean square error is less than the second set value, the control unit 11 determines that the prediction evaluation value is within the allowable range.

[0108] Here, when it is determined that the prediction evaluation value is within the allowable range, the control unit 11 determines that the prediction model 123 has been learned to the extent that it can execute a high-precision prediction process, stores the learned prediction model 123 in the storage unit 12, and enables the prediction process by the prediction model 123 (S17).

[0109] On the other hand, in step S203, when it is determined that the prediction evaluation value is outside the allowable range, the control unit 11 determines, in the next step S204, whether or not the number of execution times of the optimization process in step S12 has reached a predetermined set number of times. In step S204, when it is determined that the number of execution times is less than the set number of times, the control unit 11 returns to step S12 and repeatedly executes the processes after step S12. That is, the control unit 11 repeatedly executes the processes after step S12 until it is determined in step S203 that the prediction evaluation value is within the allowable range.

[0110] On the other hand, in step S204, when it is determined that the number of executions is equal to or greater than the set number, the control unit 11 determines that the prediction model 123 has been learned to the extent that it can execute high-precision prediction processing, stores the learned prediction model 123 in the storage unit 12, and enables the prediction processing by the prediction model 123 (S17). That is, the control unit 11 repeatedly executes the processing after step S12 until it is determined in step S204 that the number of executions is equal to or greater than the set number.

[0111] [Material Prediction Processing] Hereinafter, with reference to the flowchart of FIG. 9, an example of the procedure of the material prediction processing executed in the material prediction system 100 will be described together with the material information prediction method of the present disclosure. Note that one or more steps included in the material prediction processing described below may be appropriately omitted, and the execution order of each step may be different as long as the same operational effects are produced. Further, in the following description, the execution subject of each step is described as the control unit 11, but the execution subject of each step may be the prediction processing unit 114 or one or more processors included in the prediction processing unit 114.

[0112] When an execution instruction for executing the material prediction processing is input, in step S31, the control unit 11 acquires the physical property information of the prediction target composition whose material and blending ratio are unknown. Note that step S31 is an example of the data acquisition step of the present disclosure.

[0113] In the next step S32, the control unit 11 performs a process of predicting the identification code and the blending information of the prediction target composition based on the physical property information acquired in step S31. Specifically, the control unit 11 inputs the physical property information acquired in step S31 to the input unit of the learned prediction model 123, causes the prediction model 123 to predict the identification code and the blending information of the prediction target composition, and outputs the identification code and the blending information of the prediction target composition from the output unit of the prediction model 123 as a prediction result. Note that step S32 is an example of the prediction step of the present disclosure.

[0114] In the next step S33, the control unit 11 collates the identification code obtained as the prediction result with the material classification list 32, and extracts the similar materials to which the same identification code as the identification code is assigned from the material classification list 32.

[0115] Thereafter, in order for the control unit 11 to display the identification code and the compounding information obtained as the prediction result, and the identification information of one or more of the similar materials extracted in step S33 on the display unit of the information terminal 20 or other external devices, it outputs to the information terminal 20 or other external devices (S34). Note that steps S33 and S34 are examples of the output steps of the present disclosure.

[0116] As described above, in the present embodiment, since the prediction model construction process described above is executed in the material prediction system 100, even for a rubber composition whose material and compounding ratio of the material are unknown, if the physical property information of the rubber composition is known, it is possible to construct a prediction model 123 that can predict the material used in the production of the rubber composition and the compounding ratio of the material from the physical property information.

[0117] In addition, since the processes after the optimization process (S12) are repeatedly executed until the prediction evaluation value falls within the allowable range, a high-quality learning dataset 33 for improving the prediction accuracy of the prediction model 123 is generated.

[0118] Furthermore, since the processes after the optimization process (S12) are repeatedly executed until the number of executions reaches the set number or more, a high-quality learning dataset 33 for further improving the prediction accuracy of the prediction model 123 is generated.

[0119] In addition, in the above-described embodiment, since the material prediction process (see FIG. 9) described above is executed in the material prediction system 100, even for a rubber composition with unknown material and mixing ratio of the material, by inputting the physical property information of the rubber composition into the prediction model 123, a highly accurate prediction process by the prediction model 123 is executed. As a result, it becomes possible to accurately predict the material used in the production of the rubber composition and the mixing ratio of the material. In addition, it is possible to efficiently predict and output the identification information of one or more of the similar materials by one prediction process. As a result, since it is not necessary to perform a plurality of prediction processes to obtain a plurality of prediction results as in the prior art, information regarding one or more similar materials can be predicted in a short time. Further, when the identification information of a plurality of the similar materials is obtained, the developer of the rubber composition can select the optimal material that is most advantageous for each condition such as cost, availability, and distribution from among them.

[0120] In the above-described embodiment, the configuration in which the prediction device 10 is provided with the data acquisition processing unit 111, the prediction processing unit 114, the similar material extraction unit 115, the output processing unit 116, and the prediction model construction unit 200 has been illustrated. However, the material prediction system 100 of the present disclosure is not limited to such a configuration. For example, the prediction model construction unit 200 may not be provided in the prediction device 10. In this case, the present disclosure may be configured as a prediction model construction device or a prediction model construction system including each processing unit of the prediction model construction unit 200. Further, the present disclosure may be configured as a learning data set generation device or a learning data set generation system including each processing unit of the prediction model construction unit 200.

[0121] The embodiments of the present disclosure described above include the following disclosure items (1) to (7).

[0122] The material information prediction method according to disclosure item (1) is a material information prediction method for predicting material information regarding a material used in the production of a composition based on physical property values of the composition, Referring to a first dataset including property information of each of a plurality of compositions, material identification information indicating materials used in the production of the compositions, and compounding information regarding the compounding of a plurality of materials, a search step of searching for a plurality of similar material identification information corresponding to the plurality of property information approximating each other by a predetermined optimization process; An assignment step of assigning an identifier common to the plurality of searched similar material identification information; Inputting the physical property value of the prediction target composed of the composition into a material prediction model learned using a second dataset for learning in which the similar material identification information in the first dataset is replaced with the identifier, and outputting, from the material prediction model, the identifier corresponding to the material of the prediction target and the compounding information of the material to which the identifier is assigned.

[0123] Disclosure item (2) is in the material information prediction method described in the above disclosure item (1), Further including an output step of extracting and outputting a plurality of the similar material identification information to which the identifier predicted by the prediction step is assigned from a material classification list in which the similar material identification information and the identifier are associated.

[0124] Disclosure item (3) is in the material information prediction method of the above disclosure item (1) or (3), Further including a data acquisition step of acquiring the physical property information of the prediction target, The prediction step inputs the physical property information acquired by the data acquisition step into the material prediction model, and outputs the identifier corresponding to the prediction target and the compounding information of the material to which the identifier is assigned from the material prediction model.

[0125] The material information prediction system according to disclosure item (4) is A material information system that predicts material information regarding materials used in the production of a composition based on the physical property value of the composition, Referring to a first data set including property information of each of a plurality of compositions, material identification information indicating a material used in the production of the composition, and compounding information regarding the compounding of a plurality of materials, a search processing unit searches for a plurality of similar material identification information corresponding to the plurality of property information that approximate each other by a predetermined optimization process. An assignment processing unit assigns an identifier common to the plurality of searched similar material identification information. Using a second data set for learning in which the similar material identification information in the first data set is replaced with the identifier, the physical property value of the prediction target composed of the composition is input to a material prediction model learned by machine learning, and the identifier corresponding to the material of the prediction target and the compounding information of the material to which the identifier is assigned are output from the material prediction model. It is provided with a prediction processing unit for outputting.

[0126] Disclosure item (5) is in the material information prediction system described in the above disclosure item (4). It further includes an output processing unit that extracts and outputs a plurality of the similar material identification information to which the identifier predicted by the prediction processing unit is assigned from a material classification list in which the similar material identification information and the identifier are associated.

[0127] Disclosure item (6) is in the material information prediction system described in the above disclosure item (4) or (5). It further includes a data acquisition unit that acquires the property information of the prediction target, The prediction processing unit inputs the property information acquired by the data acquisition unit into the material prediction model, and outputs the identifier corresponding to the prediction target and the compounding information of the material to which the identifier is assigned from the material prediction model.

[0128] The prediction model construction method according to disclosure item (7) is a method for constructing a prediction model for predicting material information regarding a material used in the production of a composition. Referring to a first dataset including property information of each of a plurality of compositions, material identification information indicating materials used in the production of the compositions, and compounding information regarding the compounding of the plurality of materials, a search step of searching for a plurality of similar material identification information corresponding to the plurality of pieces of property information whose property values are approximated to each other by a predetermined optimization process; An assignment step of assigning an identifier common to the plurality of searched similar material identification information; A learning dataset generation step of generating a second dataset for learning for training the prediction model by substituting the similar material identification information in the first dataset with the identifier; A learning step of performing machine learning on the prediction model using the second dataset, and includes.

Explanation of Signs

[0129] 10: Prediction device 11: Control unit 12: Storage unit 20: Information terminal 30: Database 31: Reference dataset 32: Material classification list 33: Learning dataset 100: Material prediction system 111: Data acquisition processing unit 114: Prediction processing unit 115: Similar material extraction unit 116: Output processing unit 121: Control program 123: Prediction model 124: Optimization model 200: Prediction model construction unit 201: Condition setting unit 202: Optimization processing unit 203: Identification code assignment unit 204: Learning dataset generation unit 205: Evaluation determination unit

Claims

1. A material information prediction method for predicting material information regarding a material used in manufacturing the composition based on physical property values of the composition, comprising: a search step of referring to a first dataset including physical property information including physical property values of a plurality of compositions, material identification information indicating the material used in manufacturing the composition, and blending information regarding blending of a plurality of materials, and searching for a plurality of similar material identification information corresponding to the plurality of physical property information approximating each other by a predetermined optimization process; an assignment step of assigning an identifier common to the plurality of searched similar material identification information; an input step of inputting the physical property value of a prediction target composed of the composition into a material prediction model machine-learned using a second dataset for learning in which the similar material identification information in the first dataset is replaced with the identifier, and outputting from the material prediction model the identifier corresponding to the material of the prediction target and the blending information of the material to which the identifier is assigned.

2. The material information prediction method according to claim 1, further comprising an output step of extracting and outputting a plurality of the similar material identification information to which the identifier predicted in the prediction step is assigned from a material classification list in which the similar material identification information and the identifier are associated.

3. further comprising a data acquisition step of acquiring the physical property information of the prediction target, The prediction step inputs the physical property information acquired in the data acquisition step into the material prediction model, and outputs from the material prediction model the identifier corresponding to the prediction target and the blending information of the material to which the identifier is assigned. The material information prediction method according to claim 1 or 2.

4. A material information system for predicting material information regarding a material used in manufacturing the composition based on physical property values of the composition, comprising: Referring to a first data set including property information of each of a plurality of compositions, material identification information indicating a material used in the production of the composition, and compounding information regarding the compounding of a plurality of materials, a search processing unit searches for a plurality of similar material identification information corresponding to the plurality of property information approximating each other by a predetermined optimization process. An assignment processing unit assigns an identifier common to the plurality of searched similar material identification information. Using a second data set for learning in which the similar material identification information in the first data set is replaced with the identifier, the physical property value of the prediction target composed of the composition is input to a material prediction model learned by machine learning, and the identifier corresponding to the material of the prediction target and the compounding information of the material to which the identifier is assigned are output from the material prediction model. A material information prediction system comprising a prediction processing unit.

Images