Estimation method, model generation method, estimation device, model generation device, and program

Abstract

To provide a technique capable of estimating characteristics of a resin composition with high accuracy.SOLUTION: The estimation method is executed by one or more computers. The estimation method according to the present embodiment includes an estimation step S11. In the estimation step S11, one or more characteristic values of the polymer composition obtained in accordance with the formulation are estimated using the formulation and the feature quantity. The blending data is data indicating a blending ratio of each of a plurality of materials including a resin and a filler. The feature amount is a feature amount relating to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the compound data, and the elastic modulus of the resin composition calculated using the compound data.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an estimation method, a model generation method, an estimation device, a model generation device, and a program. [Background technology]

[0002] When attempting to obtain a resin composition with desired properties, if it were possible to accurately estimate the properties of the resin composition that can be obtained from the blend of materials used to obtain the resin composition, the labor required for experiments could be reduced. Technology has been developed that uses computer-based estimation of the properties of the resin composition obtained from the blend based on information about the blend.

[0003] Patent Document 1 describes that a computer executes a process of inputting formulation information into a trained model and estimating physical property information of an adhesive corresponding to the input formulation information. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-26478 Summary of the Invention [Problem to be solved by the invention]

[0005] However, there is room for improvement in the accuracy of estimation of the properties of a resin composition.

[0006] One aspect of the present invention provides a technique that enables the properties of a resin composition to be estimated with high accuracy. [Means for solving the problem]

[0007] According to one embodiment of the present invention, there are provided the following estimation method, model generation method, estimation device, model generation device, and program.

[0008] 1. A method of estimation implemented by one or more computers, comprising: an estimation step of estimating one or more characteristic values ​​of a resin composition obtained according to blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and feature amounts; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Estimation method. 2. In the estimation method described in 1., The method further includes a feature amount specifying step of specifying the feature amount using the blending data. Estimation method. 3. In the estimation method according to 1. or 2., The characteristic amount includes a value obtained by multiplying the particle size of the filler by the specific surface area of ​​the filler. Estimation method. 4. In the estimation method according to any one of 1. to 3., The one or more characteristic values ​​include one or more of specific gravity, water absorption, diffusion coefficient, volume resistivity, electrical conductivity, surface resistance, refractive index, relative dielectric constant, total light transmittance, glass transition temperature, linear expansion coefficient, volume expansion coefficient, thermal conductivity, thermal diffusivity, specific heat, die shear strength, peel strength, flexural strength, tensile strength, compressive strength, dynamic modulus, tensile modulus, flexural modulus, elongation, and viscosity. Estimation method. 5. In the estimation method according to any one of 1. to 4., The method further includes a clustering step of clustering the plurality of blending data into a plurality of clusters. Estimation method. 6. In the estimation method described in 5., The method further includes a representative combination data extraction step of extracting one or more combination data from each of the plurality of clusters. Estimation method. 7. In the estimation method according to any one of 1. to 6., In the estimation step, a plurality of characteristic values ​​of the resin composition are estimated, The method further includes a preference calculation step of calculating a preference for the combination data using the plurality of characteristic values ​​and weighting coefficients for each of the plurality of characteristic values. Estimation method. 8. In the estimation method described in 7., a display data output step of outputting display data for displaying the weighting coefficient input screen; a weighting coefficient acquisition step of acquiring the weighting coefficient input on the input screen; In the preference calculation step, the preference is calculated using the weighting coefficient acquired in the weighting coefficient acquisition step. Estimation method. 9. In the estimation method according to 7. or 8., The method further includes a blending output step of outputting the blending data and the calculated preference. Estimation method. 10. In the estimation method according to 7. or 8., In the estimation step, the plurality of characteristic values ​​are estimated for each of the plurality of blending data, The method further includes a combination output step of outputting the plurality of combination data in order based on the preference. Estimation method. 11. In the estimation method according to 7. or 8., In the estimation step, the plurality of characteristic values ​​are estimated for each of the plurality of blending data, The method further includes a combination output step of outputting one or more combination data extracted from the plurality of combination data based on the preference. Estimation method. 12. In the estimation method according to any one of 1. to 11., In the estimation step, the one or more characteristic values ​​are estimated using an estimation model that has been subjected to machine learning. Estimation method. 13. An estimation unit is provided that estimates one or more characteristic values ​​of a resin composition obtained according to blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and feature quantities; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Estimation device. 14. Computer, an estimation unit that estimates one or more characteristic values ​​of a resin composition obtained according to the blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and the feature amounts; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. program. 15. A model generation method implemented by one or more computers, comprising: a model generation step of generating an estimation model by performing machine learning using training data including blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Model generation method. 16. A model generation unit is provided that generates an estimation model by performing machine learning using training data including blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Model generation device. 17. Computer, a model generation unit that generates an estimation model by performing machine learning using learning data that includes blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. program. [Effects of the Invention]

[0009] According to one aspect of the present invention, a technique can be provided that enables the properties of a resin composition to be estimated with high accuracy. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a flowchart showing an outline of an estimation method according to a first embodiment. [Figure 2] 1 is a block diagram showing an overview of an estimation device according to a first embodiment. [Figure 3] 1 is a block diagram illustrating a functional configuration of an estimation device according to a first embodiment. [Figure 4] 4 is a flowchart illustrating the flow of an estimation method according to the first embodiment. [Figure 5] FIG. 10 is a diagram illustrating a computer for realizing the estimation device. [Figure 6] FIG. 2 is a diagram illustrating an example of the configuration of combination data. [Figure 7] FIG. 10 is a diagram illustrating an estimation model used by an estimation unit. [Figure 8]1 is a flowchart showing an overview of a model generation method according to the first embodiment. [Figure 9] 1 is a block diagram illustrating a functional configuration of a model generation device according to a first embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of the functional configuration of an estimation device according to a second embodiment. [Figure 11] 10 is a flowchart illustrating the flow of an estimation method according to a second embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of the configuration of constraint information. [Figure 13] FIG. 10 is a diagram illustrating an example of the configuration of request information. [Figure 14] FIG. 10 is a diagram illustrating an example of the functional configuration of an estimation device according to a third embodiment. [Figure 15] 10 is a flowchart illustrating the flow of an estimation method according to a third embodiment. [Figure 16] FIG. 10 is a diagram illustrating an example of the functional configuration of an estimation device according to a fourth embodiment. [Figure 17] 10 is a flowchart illustrating the flow of an estimation method according to a fourth embodiment. [Figure 18] FIG. 10 is a diagram illustrating an example of the configuration of weight information. [Figure 19] 10 is a flowchart showing a modified example of the estimation device according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and their description will be omitted where appropriate.

[0012] (First embodiment) FIG. 1 is a flowchart showing an outline of an estimation method according to a first embodiment. The estimation method according to this embodiment is executed by one or more computers. The estimation method according to this embodiment includes an estimation step S11. In the estimation step S11, one or more characteristic values ​​of a resin composition obtained according to the formulation data are estimated using the formulation data and feature amounts. The formulation data is data indicating the formulation ratios of each of a plurality of materials including a resin and a filler. The feature amounts are feature amounts related to at least one of the shape of the filler, the interparticle distance of the filler in the resin composition calculated using the formulation data, and the elastic modulus of the resin composition calculated using the formulation data.

[0013] FIG. 2 is a block diagram showing an overview of the estimation device 10 according to this embodiment. The estimation device 10 according to this embodiment includes an estimation unit 130. The estimation unit 130 uses formulation data and feature quantities to estimate one or more characteristic values ​​of a resin composition obtained according to the formulation data. The formulation data is data indicating the blending ratios of a plurality of materials including a resin and a filler. The feature quantities are feature quantities related to at least one of the shape of the filler, the interparticle distance of the filler in the resin composition calculated using the formulation data, and the elastic modulus of the resin composition calculated using the formulation data.

[0014] The estimation method according to this embodiment can be executed by an estimation device 10.

[0015] One example is the design of a compound to obtain a composite material with a filler dispersed in a resin matrix for a specific application. In such compound design, it is necessary to determine the combination and compounding ratio of multiple materials while adjusting multiple physical property values ​​to meet the required characteristics. In particular, when compound design is performed including trace amounts of additives, the number of materials increases, and the cost and time required for compound design increase. Therefore, technology that can reduce the effort required for compound design is needed.

[0016] First, by accurately estimating the properties of the resin composition obtained with each formulation, it becomes easier to narrow down the candidates to those formulations that will result in a resin composition having the required properties.

[0017] In particular, when a resin composition contains a filler, the properties of the resin composition are strongly dependent on the filler content and the characteristics of the filler. Furthermore, fillers that can be used as materials come in a variety of shapes, and even if the material is the same, the properties may change if fillers with different shapes are used. Therefore, by using feature quantities that quantitatively represent the characteristics related to the filler's shape, the estimation accuracy (generalization performance) of the properties of the resulting resin composition can be improved. Furthermore, by using feature quantities related to the interparticle distance of the filler in the resin composition and the theoretical elastic modulus of the resin composition containing the filler, the estimation accuracy (generalization performance) of the properties of the resin composition can be improved.

[0018] FIG. 3 is a block diagram illustrating an example of the functional configuration of the estimation device 10 according to this embodiment. In the example of FIG. 3, the estimation device 10 further includes a feature quantity identification unit 120 and an ingredient storage unit 121. FIG. 4 is a flowchart illustrating the flow of the estimation method according to this embodiment. In the example of FIG. 4, the estimation method according to this embodiment further includes a feature quantity identification step S10. In the feature quantity identification step S10, the feature quantity identification unit 120 identifies feature quantities using blending data.

[0019] The hardware configuration of the estimation device 10 will be described below. Each functional component of the estimation device 10 (the feature identification unit 120 and the estimation unit 130) may be realized by hardware (e.g., a hardwired electronic circuit) that realizes the functional component, or by a combination of hardware and software (e.g., a combination of an electronic circuit and a program that controls it). Below, a case where each functional component of the estimation device 10 is realized by a combination of hardware and software will be further described.

[0020] FIG. 5 is a diagram illustrating a computer 1000 for realizing the estimation device 10. The computer 1000 is any computer. For example, the computer 1000 is a system on chip (SoC), a personal computer (PC), a server machine, a tablet terminal, a smartphone, or the like. The computer 1000 may be a dedicated computer designed to realize the estimation device 10, or may be a general-purpose computer. Furthermore, the estimation device 10 may be realized by a single computer 1000 or by a combination of multiple computers 1000.

[0021] The computer 1000 includes a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input / output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path through which the processor 1040, the memory 1060, the storage device 1080, the input / output interface 1100, and the network interface 1120 transmit and receive data to and from each other. However, the method of interconnecting the processor 1040 and other components is not limited to bus connection. The processor 1040 may be any of various processors, such as a central processing unit (CPU), a graphics processing unit (GPU), or a field-programmable gate array (FPGA). The memory 1060 is a main storage device implemented using a random access memory (RAM) or the like. The storage device 1080 is an auxiliary storage device implemented using a hard disk, a solid state drive (SSD), a memory card, a read-only memory (ROM), or the like.

[0022] The input / output interface 1100 is an interface for connecting the computer 1000 to an input / output device. For example, an input device such as a keyboard and an output device such as a display are connected to the input / output interface 1100. The input / output interface 1100 may be connected to the input device or output device via a wireless connection or a wired connection.

[0023] The network interface 1120 is an interface for connecting the computer 1000 to a network. This communication network may be, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). The network interface 1120 may be connected to the network via a wireless connection or a wired connection.

[0024] The storage device 1080 stores program modules that realize the respective functional components of the estimation device 10. The processor 1040 reads each of these program modules into the memory 1060 and executes them to realize the function corresponding to each program module.

[0025] The resin composition (hereinafter also referred to as "target resin composition") whose characteristic value is to be estimated by the estimation unit 130 is not particularly limited, but can be, for example, a paste-like composition or a liquid composition. The use of the target resin composition is not particularly limited, but the target resin composition is, for example, a die bonding paste, an ignition coil insulation casting composition, a relay insulation sealing composition (for automotive, communication, power, industrial, etc.), or an electronic product insulation adhesive composition.

[0026] The one or more characteristic values ​​estimated by the estimation unit 130 are not particularly limited, and examples of the one or more characteristic values ​​estimated by the estimation unit 130 include one or more of specific gravity, water absorption, volume resistivity, electrical conductivity, surface resistance, refractive index, relative dielectric constant, total light transmittance, glass transition temperature, linear expansion coefficient, volume expansion coefficient, thermal conductivity, thermal diffusivity, specific heat, die shear strength, peel strength, flexural strength, tensile strength, compressive strength, dynamic modulus, tensile modulus, flexural modulus, elongation, and viscosity. The number of characteristic values ​​estimated by the estimation unit 130 is not particularly limited, and may be one, two or more, or three or more.

[0027] FIG. 6 is a diagram illustrating the configuration of formulation data. The formulation data indicates the formulation ratio of each of multiple materials. The multiple materials (hereinafter also referred to as a "material group") whose formulation ratios are indicated in the formulation data include at least a resin and a filler. The material group may include multiple resins that are different from each other. The material group may also include multiple fillers that are different from each other. The materials included in the material group may be identified by product number or the like. The formulation ratio of each material indicated in the formulation data may be, for example, a value indicating the volume ratio or mass ratio of the material in the target resin composition.

[0028] The resin is not particularly limited, but for example, the material group can include one or more resins selected from the group consisting of epoxy resin, phenolic resin, melamine resin, unsaturated polyester resin, divinylbenzene polymer, divinylbenzene-styrene copolymer, divinylbenzene-acrylic acid ester copolymer, diallyl phthalate polymer, triallyl isocyanurate polymer, and benzoguanamine polymer.

[0029] The filler is not particularly limited, and for example, the group of materials may include one or more fillers selected from the group consisting of conductive particles and non-conductive particles. Examples of conductive particles include silver, gold, copper, silica, alumina, and organic fillers. Examples of non-conductive particles include glass. The group of materials may include one or more fillers selected from the group consisting of silver, gold, copper, silica, alumina, organic fillers, and glass.

[0030] The group of materials may include one or more other materials selected from the group consisting of a curing agent, a coupling agent, a curing accelerator, a diluent, a radical initiator, and a stress reducing agent. The group of materials may or may not include a solvent. The group of materials may also include a composition as a material.

[0031] As described above, the feature quantity used by the estimation unit 130 according to this embodiment relates to at least one of the shape of the filler, the interparticle distance of the filler in the target resin composition, and the elastic modulus of the target resin composition. The feature quantity may be a scalar or a vector combining multiple numerical values.

[0032] Examples of feature quantities related to the shape of the filler will be described. Feature quantities related to the shape of the filler include, for example, a shape factor. The shape factor is a value obtained by multiplying the particle size of the filler by the specific surface area of ​​the filler. The particle size D50 at a passing mass percentage of 50% can be used as the particle size of the filler.

[0033] However, the feature quantity related to the shape of the filler is not limited to the shape factor. The feature quantity related to the shape of the filler can include one or more of the particle diameter D50 at a mass percentage of 50%, the shape factor, the equivalent circle diameter, the major axis diameter, the minor axis diameter, the maximum major axis diameter, the particle perimeter, the equivalent sphere volume, the circularity, the perimeter envelope, and the aspect ratio. The equivalent circle diameter, the major axis diameter, the minor axis diameter, the maximum major axis diameter, the particle perimeter, the equivalent sphere volume, the circularity, the perimeter envelope, and the aspect ratio are each shape characteristic values ​​obtained by analyzing captured images of the particles. Specifically, the feature quantity can include one or more of the peak value, the average value, and the median value in a set of shape characteristic values ​​obtained by capturing images of the filler and analyzing the images of a predetermined number of filler particles.

[0034] The meaning of each shape characteristic value is explained below. The equivalent circle diameter is the diameter of a circle with the same area as the imaged particle. The minor axis diameter is the smallest distance between two pairs of parallel lines tangent to the particle in the image. The direction perpendicular to the parallel lines at this smallest distance is called the minor axis. The major axis diameter is the length of the particle in the direction perpendicular to the minor axis (the major axis). The maximum major axis diameter is the maximum distance between two points on the particle's outline in the image. The particle perimeter is the circumference of the particle. The equivalent sphere volume is calculated by (π × the cube of the equivalent circle diameter) ÷ 6. Circularity is the ratio of the circumference of a circle with the same area as the imaged particle to the particle perimeter. For example, if the imaged particle is a perfect circle, the circularity is 1. The perimeter envelopment is calculated by dividing the convex hull perimeter by the particle's actual perimeter. For example, the smoother the outline of the imaged particle, the closer the perimeter envelopment is to 1. The aspect ratio is the ratio of the major axis diameter to the minor axis diameter of a particle.

[0035] An example of a feature related to the interparticle distance of filler particles in a target resin composition is described below. In a resin composition, the resin and filler typically have different properties. For example, the thermal conductivity of the resin is low, while the thermal conductivity of the filler is high. As a result, heat is transmitted through the filler within the resin composition. Therefore, by using the thickness of the resin between filler particles, i.e., the interparticle distance, as a feature, the estimation accuracy (generalization performance) of thermal conductivity can be improved. A similar concept can be generalized and applied to the estimation of other physical properties, such as the way force and electricity are transmitted.

[0036] The distance between filler particles in the target resin composition is, for example, the particle diameter d p and the blending ratio of the filler. Specifically, the interparticle distance h can be calculated using the following formula (1) described in "Basic Overview of Microparticle Dispersion in Liquid" (Kamiya Hidehiro, Journal of the Japan Society of Color Materials, 2013, Vol. 86, No. 1, pp. 26-30). F is the volume concentration of the filler in the target resin composition, and can be calculated using the blending ratio of the filler. The particle diameter d of the filler pFor example, the particle size D50 at a passing mass percentage of 50% can be used as the particle size.

[0037]

number

[0038] An example of a feature related to the elastic modulus of a target resin composition is described below. The elastic modulus of a resin composition depends on the filler and resin materials and their blending ratios. The shape of the filler also affects the way force is transmitted, which can change the elastic modulus of the entire resin composition. Therefore, the elastic modulus quantitatively reflects the filler and resin materials, their blending ratios, and the filler shape. Therefore, using the elastic modulus as a feature is effective when estimating other physical properties. Therefore, using a theoretically calculated elastic modulus as a feature can improve the estimation accuracy (generalization performance) of the properties of the target resin composition. Examples of elastic modulus include dynamic elastic modulus, tensile elastic modulus, and flexural elastic modulus.

[0039] The theoretical elastic modulus of the target resin composition can be calculated, for example, using the elastic modulus of the filler, the elastic modulus of the resin, the blending ratio of the filler and the resin, and information on the shape of the filler. Specifically, for example, the theoretical elastic modulus E of the target resin composition can be calculated based on the following formulas (2) to (4) described in "Mechanical Properties of Polymer Nanocomposites Filled with High Aspect Ratio Fillers" (Moriya (Morimune) et al., Journal of the Adhesion Society of Japan, 2017, Vol. 53, No. 10, pp. 348-354). c can be calculated.

[0040]

number

number

number

[0041] In addition, V f is the volumetric filling rate of the filler (vol%), and E m is the elastic modulus of the resin, and E f is the elastic modulus of the filler. If the filler is fibrous, ξ=2×(l f / t f ) and when the filler is plate-like, ξ=(2 / 3)×(l f / t f ) and when the filler is spherical or cubic, ξ=2. Here, l f is the major axis diameter of the filler, and t f is the diameter of the filler when the filler is fibrous, and is the thickness of the filler when the filler is plate-like.

[0042] Even when the elastic modulus of a target resin composition is to be estimated, the estimation accuracy can be improved by using the theoretical elastic modulus as a feature.

[0043] The material storage unit 121 stores information about materials included in the material group in advance. In the feature quantity identification step S10, the feature quantity identification unit 120 can identify feature quantities using the blending data and the information about the materials stored in the material storage unit 121. The material storage unit 121 may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When the material storage unit 121 is provided within the estimation device 10, the material storage unit 121 is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0044] In the estimation step S10, the estimation unit 130 can estimate one or more characteristic values ​​using an estimation model that has undergone machine learning.

[0045] 7 is a diagram illustrating an estimation model 131 used by the estimation unit 130. The estimation model 131 receives at least input of formulation data and feature quantities. The estimation model 131 then outputs estimated characteristic values ​​(hereinafter also referred to as "estimated characteristic values") of a target resin composition corresponding to the formulation data. The target resin composition corresponding to certain formulation data refers to a resin composition obtained by mixing materials included in a material group according to the blending ratios indicated in the formulation data.

[0046] When the estimation unit 130 estimates multiple characteristic values ​​for one blending data, the estimation unit 130 may use an estimation model 131 prepared for each characteristic value, or may use an estimation model 131 that can output multiple estimated characteristic values.

[0047] The estimation model 131 can estimate the characteristic values ​​of the target resin composition with high accuracy by using the feature amounts described above.

[0048] The flow of the estimation method according to this embodiment will be described in detail below with reference to FIGS.

[0049] In one example, the feature quantity identification unit 120 of the estimation device 10 acquires combination data. For example, the feature quantity identification unit 120 can acquire the combination data by reading out the combination data stored in advance in a storage unit accessible by the feature quantity identification unit 120. This storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. If the storage unit is provided within the estimation device 10, the storage unit is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the estimation device 10. Alternatively, the feature quantity identification unit 120 may acquire the combination data from a device other than the estimation device 10 or from another functional component within the estimation device 10. Alternatively, the feature quantity identification unit 120 may acquire the combination data by having a user input the combination data into the estimation device 10 using an input device or the like, and the feature quantity identification unit 120 accepts the input combination data.

[0050] In the feature quantity identification step S10, the feature quantity identification unit 120 identifies the materials to be blended into the target resin composition (i.e., materials whose blending ratio is not zero) based on the acquired blending data. Then, the feature quantity identification unit 120 reads information about the identified materials from the material storage unit 121 and identifies the feature quantities using that information. Here, the feature quantity identification unit 120 further identifies the feature quantities using the blending ratios of each material as necessary.

[0051] For example, the feature quantity specifying unit 120 can read out a shape coefficient, a circle equivalent diameter, a major axis diameter, or the like from the material storage unit 121 and use these as feature quantities related to the shape of the filler. p The feature quantity specifying unit 120 can read out the elastic modulus of the filler, the elastic modulus of the resin, and the aspect ratio of the filler from the material storage unit 121, and calculate the theoretical elastic modulus using the read-out information and the filler / resin blending ratio.

[0052] When a plurality of fillers are blended into the target resin composition as mutually different materials, a feature quantity for all fillers contained in the target resin composition can be calculated based on the blending ratio of the plurality of fillers. Specifically, for example, the feature quantity identifying unit 120 can calculate a weighted average of the feature quantities (e.g., circle equivalent diameters) related to the shapes of the plurality of fillers based on the blending ratios of the plurality of fillers as a feature quantity (e.g., circle equivalent diameter) related to the shape of the entire filler, and include this in the feature quantities used for estimation.

[0053] The estimation unit 130 acquires the combination data and the feature amount generated by the feature amount identification unit 120 in association with each other.

[0054] In the estimation step S11, the estimation unit 130 can read and use an estimation model 131 stored in advance in a storage unit accessible from the estimation unit 130. This storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When this storage unit is provided within the estimation device 10, this storage unit is realized using, for example, a storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0055] The estimation unit 130 inputs the combination data and the feature quantities into the estimation model 131 and causes the estimation model 131 to output estimated characteristic values. The estimation unit 130 outputs the obtained one or more estimated characteristic values ​​as estimation results for the combination data. The estimation unit 130 may output the combination data and the estimation results for the combination data in association with each other.

[0056] In this embodiment, the estimation unit 130 may acquire multiple combination data and estimate one or more characteristic values ​​for each of the multiple combination data. In this case, the estimation unit 130 may output the estimation results for the multiple combination data one by one, or may output multiple estimation results collectively.

[0057] The estimation unit 130 stores the estimation result in, for example, a storage unit accessible from the estimation unit 130. The storage unit may be provided within the estimation device 10 or may be provided outside the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized using, for example, a storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0058] As another example, the estimation unit 130 may output the estimation result to another functional configuration unit within the estimation device 10, or to a device other than the estimation device 10. Alternatively, the estimation unit 130 may display the estimation result on a display connected to the computer 1000 that realizes the estimation device 10.

[0059] Although the example in which the feature quantity identification unit 120 identifies the feature quantities has been described above, the estimation device 10 does not necessarily have to include the feature quantity identification unit 120 and the ingredient storage unit 121. In this case, the estimation unit 130 can acquire the blending data and feature quantities by, for example, reading them from a storage unit accessible from the estimation unit 130. This storage unit may be provided within the estimation device 10 or external to the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized, for example, by using the storage device 1080 of the computer 1000 that implements the estimation device 10. Alternatively, the estimation unit 130 may acquire the blending data and feature quantities from a device other than the estimation device 10 or from another functional component within the estimation device 10. Alternatively, the user may input the blending data and feature quantities to the estimation device 10 using an input device or the like, and the estimation unit 130 may accept the input information and acquire the blending data and feature quantities.

[0060] FIG. 8 is a flowchart showing an outline of a model generation method according to this embodiment. The model generation method according to this embodiment is executed by one or more computers. The model generation method according to this embodiment includes a model generation step S21. In the model generation step S21, an estimated model 131 is generated by performing machine learning using training data. The training data includes formulation data, feature values, and one or more characteristic values ​​of the resin composition obtained according to the formulation data. The formulation data indicates the formulation ratio of each of multiple materials including the resin and the filler. The feature values ​​are feature values ​​related to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the formulation data, and the elastic modulus of the resin composition calculated using the formulation data.

[0061] The estimation model generated by the model generation method according to this embodiment can be used as the estimation model 131 in the estimation device 10.

[0062] FIG. 9 is a block diagram illustrating the functional configuration of a model generation device 20 according to this embodiment. The model generation device 20 according to this embodiment includes a model generation unit 210. The model generation unit 210 generates an estimated model by performing machine learning using training data. The training data includes formulation data, feature values, and one or more characteristic values ​​of the resin composition obtained according to the formulation data. The formulation data indicates the formulation ratio of each of multiple materials including a resin and a filler. The feature values ​​are feature values ​​related to at least one of the shape of the filler, the interparticle distance of the filler in the resin composition calculated using the formulation data, and the elastic modulus of the resin composition calculated using the formulation data.

[0063] The model generation method according to this embodiment can be executed by a model generation device 20.

[0064] According to the model generation method and model generation device 20 of this embodiment, an estimation model 131 is obtained that can estimate characteristics with high accuracy using feature quantities related to at least one of the shape of the filler, the distance between filler particles, and the elastic modulus.

[0065] The hardware configuration of a computer that realizes the model generation device 20 according to this embodiment is shown in FIG. 5 , similarly to the estimation device 10. However, a storage device 1080 of a computer 1000 that realizes the model generation device 20 according to this embodiment stores program modules that realize each functional component (model generation unit 210) of the model generation device 20. A processor 1040 of the computer 1000 that realizes the model generation device 20 reads each of these program modules into a memory 1060 and executes them to realize the function corresponding to each program module. Furthermore, the model generation device 20 may be realized by a single computer 1000 or by a combination of multiple computers 1000. One or more computers 1000 that realize the model generation device 20 may also serve as one or more computers 1000 that realize the estimation device 10.

[0066] The combination data and feature amounts included in the training data are as described above.

[0067] One or more characteristic values ​​(hereinafter also referred to as "correct characteristic values") included in the learning data are used as correct data in machine learning. Examples of correct characteristic values ​​are as exemplified as one or more characteristic values ​​estimated by the estimation unit 130. Correct characteristic values ​​are obtained, for example, by measuring the properties of a resin composition prepared in advance according to formulation data. A resin composition prepared according to formulation data means a resin composition obtained by mixing materials included in a material group according to the formulation ratio indicated in the formulation data.

[0068] The flow of the model generation method according to this embodiment will be described in detail below. In one example, the model generation unit 210 of the model generation device 20 acquires training data. For example, the model generation unit 210 can acquire the training data by reading out training data previously stored in a storage unit accessible by the model generation unit 210. This storage unit may be provided within the model generation device 20 or may be provided externally to the model generation device 20. When this storage unit is provided within the model generation device 20, this storage unit is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the model generation device 20. Alternatively, the model generation unit 210 may acquire the training data from a device other than the model generation device 20, or from another functional configuration unit within the model generation device 20.

[0069] As another example, the model generation unit 210 may acquire a pair of combination data and correct characteristic values, and identify features for the combination data in the same manner as performed by the feature identification unit 120. The model generation unit 210 can generate training data by combining the identified features, the combination data, and the correct characteristic values. The model generation unit 210 can acquire information necessary for generating training data by reading it from a storage unit accessible by the model generation unit 210, such as that described above.

[0070] The model generation unit 210 can perform machine learning using existing technology. Specifically, the model generation unit 210 inputs the combination data and feature quantities included in the learning data into the estimation model 131. Then, the model generation unit 210 updates multiple parameters for configuring the estimation model 131 so as to reduce the difference between the estimated characteristic value output from the estimation model 131 and the correct characteristic value included in the learning data.

[0071] The model generation unit 210 similarly acquires multiple pieces of training data and repeats parameter updates until a termination condition is met, such as when a predetermined number of parameter updates have been performed, or when the difference between the estimated characteristic value and the correct characteristic value included in the training data falls below a predetermined threshold.

[0072] The model generation unit 210 can store the trained estimation model 131 in a storage unit accessible from the model generation unit 210. This storage unit may be provided inside the model generation device 20 or may be provided outside the model generation device 20. When this storage unit is provided inside the model generation device 20, this storage unit is realized using, for example, the storage device 1080 of the computer 1000 that realizes the model generation device 20.

[0073] According to this embodiment, highly accurate characteristic estimation is possible by using feature quantities related to at least one of the shape of the filler, the interparticle distance of the filler within the resin composition calculated using the formulation data, and the elastic modulus of the resin composition calculated using the formulation data.

[0074] (Second embodiment) 10 is a diagram illustrating a functional configuration of an estimation device 10 according to a second embodiment. The estimation device 10 according to this embodiment is the same as the estimation device 10 according to the first embodiment, except that it further includes a combination data generation unit 110 and an extraction unit 150.

[0075] 11 is a flowchart illustrating the flow of the estimation method according to the present embodiment. The estimation method according to the second embodiment includes a blending data generation step S30, a feature quantity identification step S31, an estimation step S32, and an extraction step S33. The feature quantity identification step S31 and the estimation step S32 according to the second embodiment are the same as the feature quantity identification step S10 and the estimation step S11 according to the first embodiment, respectively.

[0076] The estimation method according to the second embodiment is the same as the estimation method according to the first embodiment, except that it includes a combination data generation step S30 and an extraction step S33. In the combination data generation step S30, the combination data generation unit 110 acquires constraint condition information related to the combination and generates multiple combination data based on the constraint condition information.

[0077] In the estimation step S32 according to this embodiment, the estimation unit 130 acquires multiple combination data generated by the combination data generation unit 110, and estimates one or more characteristic values ​​for each combination data, as described in the first embodiment. In the extraction step S33, the extraction unit 150 extracts combination data from the multiple combination data in which the estimated one or more characteristic values ​​satisfy the required characteristics. The estimation device 10 and estimation method according to this embodiment will be described in detail below.

[0078] The hardware configuration of the computer that realizes the estimation device 10 according to this embodiment is shown in Fig. 5, for example, similar to the estimation device 10. However, the storage device 1080 of the computer 1000 that realizes the estimation device 10 according to this embodiment further stores program modules that realize the functions of the blending data generation unit 110 and the extraction unit 150.

[0079] 12 is a diagram illustrating an example of the configuration of constraint information. The constraint information is information that indicates constraints related to blending. For example, the constraint information indicates the range of allowable blending ratios for each material included in a material group.

[0080] In the combination data generation step S30, the combination data generation unit 110 of the estimating device 10 can acquire the constraint condition information by reading it from a storage unit accessible from the combination data generation unit 110. This storage unit may be provided within the estimating device 10 or may be provided externally to the estimating device 10. When the storage unit is provided within the estimating device 10, the storage unit is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the estimating device 10. Alternatively, the combination data generation unit 110 may acquire the constraint condition information from a device other than the estimating device 10, or from another functional component within the estimating device 10. Alternatively, the user may input the constraint condition information to the estimating device 10 using an input device or the like, and the input constraint condition information may be received by the combination data generation unit 110.

[0081] The combination data generation unit 110 generates combination data that satisfies the constraint conditions indicated in the acquired constraint condition information. That is, the combination data generation unit 110 specifies the combination ratio within the range indicated by the constraint conditions for each material included in the material group. In this way, the combination data generation unit 110 generates multiple combination data. The number of combination data generated by the combination data generation unit 110 may be determined in advance or may be input to the estimation device 10 by a user. The combination data generation unit 110 comprehensively generates multiple combination data while determining the combination ratio of each material at intervals corresponding to the number of combination data to be generated, for example.

[0082] In the feature quantity identification step S31, the feature quantity identification unit 120 acquires the multiple combination data generated by the combination data generation unit 110. Then, the feature quantity identification unit 120 identifies the feature quantity for each of the multiple combination data in the same manner as described in the first embodiment. The feature quantity identification unit 120 associates the feature quantity identified based on the combination data with each combination data.

[0083] In estimation step S32, the estimation unit 130 inputs each of the multiple pairs of combination data and feature quantities into the estimation model 131 and outputs one or more estimated characteristic values. In this way, one or more estimated characteristic values ​​are obtained for each of the multiple combination data. The estimation unit 130 associates one or more estimated characteristic values ​​with each combination data.

[0084] In the extraction step S33, the extraction unit 150 acquires from the estimation unit 130 a plurality of sets of combination data and one or more estimated characteristic values ​​that are associated with each other.

[0085] The extraction unit 150 also acquires requirement information indicating required characteristics. Fig. 13 is a diagram illustrating an example of the configuration of the requirement information. The requirement information indicates conditions that each of one or more estimated characteristic values ​​must satisfy.

[0086] The extraction unit 150 can acquire the request information by, for example, reading it from a storage unit accessible from the extraction unit 150. This storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the estimation device 10. Alternatively, the extraction unit 150 may acquire the request information from a device other than the estimation device 10, or from another functional configuration unit within the estimation device 10. Alternatively, the extraction unit 150 may acquire the request information by having a user input the request information to the estimation device 10 using an input device or the like, and having the extraction unit 150 accept the input request information.

[0087] The extraction unit 150 determines whether or not the one or more estimated characteristic values ​​satisfy all of the conditions indicated in the request information for each of the multiple sets of combination data and one or more estimated characteristic values ​​acquired from the estimation unit 130. Then, the extraction unit 150 extracts one or more sets in which the one or more estimated characteristic values ​​satisfy all of the conditions indicated in the request information from the multiple sets.

[0088] The extraction unit 150 outputs the one or more extracted sets. The extraction unit 150 stores the one or more extracted sets in, for example, a storage unit accessible from the extraction unit 150. The storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized using, for example, a storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0089] As another example, the extraction unit 150 may output the extracted one or more sets to another functional configuration unit within the estimating device 10, or to a device other than the estimating device 10. Alternatively, the extraction unit 150 may display the extracted one or more sets on a display connected to the computer 1000 that realizes the estimating device 10.

[0090] If there is no set among the multiple sets in which one or more estimated characteristic values ​​satisfy all of the conditions specified in the request information, the extraction section 150 may output information indicating that no set has been extracted.

[0091] According to this embodiment, the same actions and effects as those of the first embodiment can be obtained. In addition, according to this embodiment, based on constraints on the blending and requirements for the characteristics, blending data that meets those conditions can be obtained.

[0092] (Third embodiment) 14 is a diagram illustrating a functional configuration of an estimation device 10 according to a third embodiment. The estimation device 10 according to this embodiment is the same as the estimation device 10 according to at least one of the first and second embodiments, except that it further includes a clustering unit 170.

[0093] 15 is a flowchart illustrating the flow of the estimation method according to this embodiment. In the example of FIG. 15, the estimation method according to this embodiment includes a combination data generation step S40, a feature quantity identification step S41, an estimation step S42, an extraction step S43, a clustering step S44, and a representative combination data extraction step S45. The combination data generation step S40, the feature quantity identification step S41, the estimation step S42, and the extraction step S43 according to this embodiment are the same as the combination data generation step S30, the feature quantity identification step S31, the estimation step S32, and the extraction step S33 according to the second embodiment, respectively.

[0094] The estimation method according to the third embodiment is the same as the estimation method according to at least one of the first and second embodiments, except that it includes a clustering step S44 and a representative combination data extraction step S45. In the clustering step S44, the clustering unit 170 clusters multiple combination data into multiple clusters. In the representative combination data extraction step S45, the clustering unit 170 extracts one or more combination data from each of the multiple clusters. The estimation device 10 and the estimation method according to this embodiment are described in detail below.

[0095] For example, a prototype experiment may be conducted to determine whether a resin composition obtained from a set (formulation data and one or more estimated property values) output from the estimation device 10 using the formulation indicated in the formulation data actually has the estimated properties. If the extraction unit 150 extracts a large number of sets, it is laborious to conduct prototype experiments for all of the sets.

[0096] In contrast, in the estimation device 10 according to this embodiment, the clustering unit 170 clusters multiple combination data, classifying the combination data into multiple clusters based on the similarity between the combination data. Then, one or more combination data are extracted from each of the multiple clusters as representative combination data. By conducting a prototype experiment using the representative combination data extracted in this manner, it is possible to search for the desired combination data with less effort than if a prototype experiment were conducted for all combination data. Furthermore, because representative combination data is extracted from each cluster, a variety of combination data can be used as the subject of the prototype experiment without being biased toward combination data of the same type.

[0097] The hardware configuration of the computer that realizes the estimation device 10 according to this embodiment is, for example, shown in Fig. 5 , similar to the estimation device 10. However, a program module that realizes the function of the clustering unit 170 is further stored in the storage device 1080 of the computer 1000 that realizes the estimation device 10 according to this embodiment.

[0098] In the clustering step S44, the clustering unit 170 acquires the multiple sets extracted by the extraction unit 150. Then, combination data is extracted from each of the multiple sets. Clustering is performed on the set consisting of the multiple extracted combination data to generate multiple clusters. Clustering can be performed by executing existing clustering processes such as the Diana method, Ward's method, group average method, shortest distance method, centroid method, weighted average method, median method, k-means method, k-means++ method, x-means method, k-medoids method, hidden Markov model, DBSCAN, fuzzy C-means method, or NMF. Each of the multiple clusters includes two or more combination data.

[0099] In the representative combination data extraction step S45, the clustering unit 170 extracts one or more combination data as representative combination data from each of the multiple clusters. The clustering unit 170, for example, extracts a predetermined number of combination data as representative combination data from each cluster. The clustering unit 170 may extract one combination data from each of the multiple clusters, or two or more combination data from each of the multiple clusters. Alternatively, the clustering unit 170 may extract a predetermined ratio of combination data to the number of combination data included in the cluster. In this way, the clustering unit 170 can extract multiple combination data as representative combination data.

[0100] If the number of combination data included in a cluster is greater than the number of combination data to be extracted from that cluster, the clustering unit 170 extracts a portion of the combination data included in each cluster as representative combination data, and does not extract the remaining as representative combination data. In this way, it is possible to extract a number of representative combination data that is less than the number of sets extracted by the extraction unit 150.

[0101] The clustering unit 170 outputs a plurality of representative combination data (i.e., a plurality of extracted combination data). The clustering unit 170 stores the extracted plurality of representative combination data in, for example, a storage unit accessible from the clustering unit 170. The storage unit may be provided within the estimation device 10, or may be provided outside the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized, for example, using the storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0102] As another example, the clustering unit 170 may output the extracted representative combination data to other functional components within the estimation device 10, or to a device other than the estimation device 10. Alternatively, the clustering unit 170 may display the extracted representative combination data on a display connected to the computer 1000 that realizes the estimation device 10.

[0103] The clustering unit 170 may further output each representative combination data in association with one or more estimated characteristic values.

[0104] However, the clustering unit 170 according to this embodiment may not perform the representative combination data extraction step S45 and output the clustering results in the above-described manner. In this case, the clustering unit 170 outputs multiple combination data in a state in which the cluster to which each combination data belongs can be identified. In this case, the user can select combination data from each cluster.

[0105] Note that the estimation device 10 according to this embodiment may not include the combination data generation unit 110. Also, the estimation method according to this embodiment may not include the combination data generation step S40. In this case, the estimation unit 130 acquires a plurality of combination data prepared in advance and estimates characteristic values, similar to the first embodiment. Also, the estimation method according to this embodiment may not include the feature quantity identification step S41. In this case, the estimation unit 130 acquires a plurality of combination data and feature quantities prepared in advance and estimates characteristic values, similar to the first embodiment.

[0106] According to this embodiment, the same actions and effects as those of the first embodiment can be obtained. In addition, according to this embodiment, the clustering unit 170 clusters multiple combination data into multiple clusters. Therefore, it is possible to search for desired combination data while grasping the overall picture with less effort than when performing confirmation experiments on all combination data.

[0107] (Fourth embodiment) 16 is a diagram illustrating a functional configuration of an estimation device 10 according to a fourth embodiment. The estimation device 10 according to this embodiment is the same as the estimation device 10 according to at least one of the first to third embodiments, except for the points described below.

[0108] 17 is a flowchart illustrating the flow of the estimation method according to this embodiment. In the example of FIG. 17, the estimation method according to this embodiment includes a combination data generation step S50, a feature quantity identification step S51, an estimation step S52, an extraction step S53, a clustering step S54, and a representative combination data extraction step S55, a preference calculation step S56, and a combination output step S57. The combination data generation step S50, the feature quantity identification step S51, the estimation step S52, the extraction step S53, the clustering step S54, and the representative combination data extraction step S55 according to this embodiment are respectively the combination data generation step S40, the feature quantity identification step S41, the estimation step S42, the extraction step S43, the clustering step S44, and the representative combination data extraction step S45 according to the third embodiment.

[0109] The estimation method according to the fourth embodiment is the same as the estimation method according to at least one of the first to third embodiments, except for the points described below.

[0110] In the estimation step S52 according to this embodiment, the estimation unit 130 estimates multiple characteristic values ​​of the resin composition. Then, in the preference calculation step S56, the preference calculation unit 190 calculates the preference for the blending data using the multiple characteristic values ​​and weighting coefficients for each of the multiple characteristic values. By calculating the preference, it becomes easier to understand which blending data is particularly preferred, taking into account the importance of each characteristic value. The estimation device 10 and estimation method according to this embodiment will be described in detail below.

[0111] The hardware configuration of the computer that realizes the estimation device 10 according to this embodiment is shown in Fig. 5, for example, similar to the estimation device 10. However, the storage device 1080 of the computer 1000 that realizes the estimation device 10 according to this embodiment further stores a program module that realizes the function of the preference calculation unit 190.

[0112] The preference calculation unit 190 according to this embodiment acquires a plurality of representative combination data extracted by the clustering unit 170. The preference calculation unit 190 further acquires a plurality of estimated characteristic values ​​associated with each of the plurality of representative combination data.

[0113] Furthermore, the preference calculation unit 190 acquires weight information indicating a weight coefficient for each of a plurality of characteristic values. Fig. 18 is a diagram illustrating an example of the configuration of the weight information. For example, the weight information indicates a weight coefficient for viscosity, a weight coefficient for thermal conductivity, etc.

[0114] The preference calculation unit 190 can acquire the weight information by, for example, reading it from a storage unit accessible in advance from the preference calculation unit 190. This storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When this storage unit is provided within the estimation device 10, it is realized, for example, by using the storage device 1080 of the computer 1000 that realizes the estimation device 10. Alternatively, the preference calculation unit 190 may acquire the weight information from a device other than the estimation device 10, or from another functional component within the estimation device 10.

[0115] As another example, the user may input the weighting coefficient to the estimation device 10 using an input device or the like, and the input weighting coefficient may be received by the preference calculation unit 190, thereby obtaining the weighting coefficient.

[0116] FIG. 19 is a flowchart illustrating a modified example of the estimation device 10 according to the present embodiment. FIG. 19 illustrates an example in which a user inputs weighting factors to the estimation device 10. The estimation method according to this modified example further includes a display data output step S561 and a weighting factor acquisition step S562. In the display data output step S561, the preference calculation unit 190 outputs display data for displaying a weighting factor input screen. This display data then causes the weighting factor input screen to be displayed on the display connected to the estimation device 10. The user inputs weighting factors for each characteristic value on the input screen. In the weighting factor acquisition step S562, the preference calculation unit 190 acquires the weighting factors entered on the input screen. Then, in the preference calculation step S56, the preference calculation unit 190 calculates the preference using the weighting factor acquired in the weighting factor acquisition step S562. This allows the user of the estimation device 10 to arbitrarily set the importance of each characteristic value and reflect it in the preference.

[0117] In the preference calculation step S56 according to this embodiment, the preference calculation unit 190 calculates the preference A using, for example, the following formula (5): where i is the characteristic value p i is a number that identifies each item (viscosity, thermal conductivity, etc.), and w i is the characteristic value p i is the weighting factor for P i is the standardized characteristic value p i In addition, Σw i =1 holds true.

[0118]

number

[0119] The method for standardizing the characteristic value is not particularly limited, but for example, Z scaling can be used. i The average value μ i and standard deviation σ i Using P i =(p i -μ i ) / σ i After standardization, P is calculated using the formulai can be obtained.

[0120] The preference calculation unit 190 calculates the average value μ i and standard deviation σ i are the multiple characteristic values ​​p estimated by the estimation unit 130 in the estimation step S52. i Alternatively, the preference calculation unit 190 may use the mean value and standard deviation calculated using the mean value μ i and standard deviation σ i Alternatively, the preference calculation unit 190 may use the mean value and standard deviation calculated for a population of multiple characteristic values ​​stored in an existing database. The existing database may include actual measured values ​​as characteristic values. Alternatively, the preference calculation unit 190 may use the mean value μ i and standard deviation σ i A predetermined value may be used as the

[0121] The preference calculation unit 190 outputs at least one combination data. There are no particular limitations on the destination of the combination data output from the preference calculation unit 190, but for example, the preference calculation unit 190 stores the combination data in a storage unit accessible from the preference calculation unit 190. The storage unit may be provided within the estimation device 10 or may be provided externally to the estimation device 10. When the storage unit is provided within the estimation device 10, the storage unit is realized using, for example, a storage device 1080 of the computer 1000 that realizes the estimation device 10.

[0122] As another example, the preference calculation unit 190 may output the combination data to another functional component within the estimation device 10, or to a device other than the estimation device 10. Alternatively, the preference calculation unit 190 may display the combination data on a display connected to the computer 1000 that implements the estimation device 10.

[0123] An example of a method for outputting data from the preference calculation unit 190 will be further described.

[0124] In the first example, in the combination output step S57, the preference calculation unit 190 outputs the combination data and the calculated preference. The preference calculation unit 190 may further output the combination data in association with one or more corresponding estimated characteristic values. By doing so, the user can understand how promising the output combination data is. The preference calculation unit 190 may output one set of combination data and preference, or may output multiple sets of combination data and preference.

[0125] In the second example, in estimation step S52, the estimation unit 130 estimates multiple characteristic values ​​for each of the multiple combination data, and in combination output step S57, the preference calculation unit 190 outputs the multiple combination data in an order based on preference. The preference calculation unit 190 can, for example, output a table showing the multiple combination data arranged in an order based on preference. When multiple combination data are displayed on the display, the preference calculation unit 190 may display the multiple combination data one by one in an order based on preference. This allows the user to easily understand the ranking of the multiple combination data.

[0126] In the third example, in the estimation step S52, the estimation unit 130 estimates multiple characteristic values ​​for each of the multiple combination data, and in the combination output step S57, the preference calculation unit 190 outputs one or more combination data extracted from the multiple combination data based on the preference. For example, the preference calculation unit 190 extracts and outputs a predetermined number of combination data with high preference (e.g., the top three combination data) from the multiple combination data. Alternatively, the preference calculation unit 190 may extract and output combination data with preference above a predetermined threshold. Alternatively, the preference calculation unit 190 extracts and outputs a predetermined proportion of combination data with high preference (e.g., the top 10% of combination data) from the multiple combination data. This allows the user to narrow down and check promising combination data.

[0127] The preference calculation unit 190 may output data in a form that combines the first to third examples described above.

[0128] 17 and 19, an example in which both clustering and calculation of preference are performed has been described above. However, calculation of preference may be performed without clustering. That is, the extraction step S53 may be followed by the preference calculation step S56 or the display data output step S561. In this case, the preference calculation unit 190 calculates preference for the combination data extracted in the extraction step S53.

[0129] Furthermore, the estimation device 10 according to this embodiment may not include the combination data generation unit 110. Furthermore, the estimation method according to this embodiment may not include the combination data generation step S50. In this case, the estimation unit 130 acquires a plurality of combination data prepared in advance and estimates characteristic values, similar to the first embodiment. Furthermore, the estimation method according to this embodiment may not include the feature quantity identification step S51. In this case, the estimation unit 130 acquires a plurality of combination data and feature quantities prepared in advance and estimates characteristic values, similar to the first embodiment.

[0130] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0131] In addition, in the flowcharts used in the above description, multiple steps (processes) are described in order, but the order of execution of the steps performed in each embodiment is not limited to the order described. In each embodiment, the order of the steps shown in the drawings can be changed to the extent that the content is not affected. Furthermore, the above-mentioned embodiments can be combined to the extent that the content is not contradictory. [Explanation of symbols]

[0132] 10 Estimation device 20 Model generation device 110 Blending data generation unit 120 Feature Identification Unit 121 Material storage section 130 Estimation part 131 Estimation Model 150 Extraction part 170 Clustering Department 190 Preference calculation part 210 Model Generation Unit 1000 calculator 1020 Bus 1040 processor 1060 memory 1080 storage device 1100 Input / Output Interface 1120 Network Interface

Claims

1. 1. A method of estimation implemented by one or more computers, comprising: an estimation step of estimating one or more characteristic values ​​of a resin composition obtained according to blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and feature amounts; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Estimation method.

2. 2. The estimation method according to claim 1, The method further includes a feature amount specifying step of specifying the feature amount using the blending data. Estimation method.

3. 3. The estimation method according to claim 1 or 2, The characteristic amount includes a value obtained by multiplying the particle size of the filler by the specific surface area of ​​the filler. Estimation method.

4. 3. The estimation method according to claim 1 or 2, The one or more characteristic values ​​include one or more of specific gravity, water absorption, diffusion coefficient, volume resistivity, electrical conductivity, surface resistance, refractive index, relative dielectric constant, total light transmittance, glass transition temperature, linear expansion coefficient, volume expansion coefficient, thermal conductivity, thermal diffusivity, specific heat, die shear strength, peel strength, flexural strength, tensile strength, compressive strength, dynamic modulus, tensile modulus, flexural modulus, elongation, and viscosity. Estimation method.

5. 3. The estimation method according to claim 1 or 2, The method further includes a clustering step of clustering the plurality of blending data into a plurality of clusters. Estimation method.

6. 6. The estimation method according to claim 5, The method further includes a representative combination data extraction step of extracting one or more combination data from each of the plurality of clusters. Estimation method.

7. 3. The estimation method according to claim 1 or 2, In the estimation step, a plurality of characteristic values ​​of the resin composition are estimated, The method further includes a preference calculation step of calculating a preference for the combination data using the plurality of characteristic values ​​and weighting coefficients for each of the plurality of characteristic values. Estimation method.

8. 8. The estimation method according to claim 7, a display data output step of outputting display data for displaying the weighting coefficient input screen; a weighting coefficient acquisition step of acquiring the weighting coefficient input on the input screen; In the preference calculation step, the preference is calculated using the weighting coefficient acquired in the weighting coefficient acquisition step. Estimation method.

9. 8. The estimation method according to claim 7, The method further includes a blending output step of outputting the blending data and the calculated preference. Estimation method.

10. 8. The estimation method according to claim 7, In the estimation step, the plurality of characteristic values ​​are estimated for each of the plurality of blending data, The method further includes a combination output step of outputting the plurality of combination data in order based on the preference. Estimation method.

11. 8. The estimation method according to claim 7, In the estimation step, the plurality of characteristic values ​​are estimated for each of the plurality of blending data, The method further includes a combination output step of outputting one or more combination data extracted from the plurality of combination data based on the preference. Estimation method.

12. 3. The estimation method according to claim 1 or 2, In the estimation step, the one or more characteristic values ​​are estimated using an estimation model that has been subjected to machine learning. Estimation method.

13. an estimation unit that estimates one or more characteristic values ​​of a resin composition obtained according to blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and feature amounts; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Estimation device.

14. Computer, an estimation unit that estimates one or more characteristic values ​​of a resin composition obtained in accordance with blending data, using blending data indicating blending ratios of each of a plurality of materials including a resin and a filler and feature amounts; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. program.

15. 1. A model generation method executed by one or more computers, comprising: a model generation step of generating an estimation model by performing machine learning using training data including blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Model generation method.

16. a model generation unit that generates an estimation model by performing machine learning using training data that includes blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. Model generation device.

17. Computer, a model generation unit that generates an estimation model by performing machine learning using learning data that includes blending data indicating blending ratios of each of a plurality of materials including a resin and a filler, feature values, and one or more characteristic values ​​of the resin composition obtained according to the blending data; The feature amount relates to at least one of the shape of the filler, the inter-particle distance of the filler in the resin composition calculated using the blending data, and the elastic modulus of the resin composition calculated using the blending data. program.

Images