JP2024035515A - Physical quantity estimation system, approximate function generation device, physical quantity estimation device, program, recording medium, and physical quantity estimation method - Google Patents

Abstract

[Problem] To estimate the value of a physical quantity for a composite material with high accuracy. [Solution] When inputting the value of a categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of a first composite material whose physical quantity value is unknown, the value of the physical quantity for the first composite material is input. The value of the physical quantity of the first composite material is estimated by using an approximation function that outputs . [Selection diagram] Figure 8

Description

The present invention relates to a physical quantity estimation system, an approximation function generation device, a physical quantity estimation device, a program, a recording medium, and a physical quantity estimation technique, and is effective when applied to, for example, a technique for estimating the value of a physical quantity depending on the blending ratio of a resin composite material. related to technology.

Japanese Unexamined Patent Publication No. 2018-156689 (Patent Document 1) describes a technique for estimating the physical properties of a composite material by utilizing artificial intelligence.

Japanese Patent Application Publication No. 2018-156689

In recent years, composite materials have been developed that provide new performance to the characteristics of the resin itself by combining multiple types of resins and compounding agents. In this regard, in order to develop a new composite material, it is necessary to develop the material while adjusting the composition ratio of each composition until the composite material has the desired properties. For this reason, the development of composite materials requires enormous costs. Therefore, from the viewpoint of increasing the efficiency of composite material development, it is desirable to be able to estimate to some extent the physical quantities of the composite material to be tested at the experimental planning stage. However, for example, composite materials for wire coating materials include many types of compounding agents, and the values of physical quantities may vary greatly depending on the compounding composition ratio. For this reason, it is difficult to estimate the values of physical quantities for composite materials. Based on the above, there is a need for a technology that can estimate the values of physical quantities for composite materials with high accuracy.

The physical quantity estimation system in one embodiment is a physical quantity estimation system that estimates the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. This physical quantity estimation system calculates the value of the first composite material by inputting the value of the first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of the first composite material whose physical quantity value is unknown. an approximation function generation unit that generates an approximation function that outputs the value of the physical quantity for the first composite material; and a physical quantity estimation unit that estimates the value of the physical quantity for the first composite material based on the value of the first categorical variable and the approximation function. Be prepared.

The approximation function generation device in one embodiment is an approximation function generation device that becomes a component of a physical quantity estimation system that estimates the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. When this approximation function generation device inputs the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of a physical quantity of a first composite material whose physical quantity value is unknown, it generates a first composite material. The apparatus includes an approximation function generation unit that generates an approximation function that outputs a value of a physical quantity for a material.

A program in one embodiment is a program that causes a computer to execute a process of estimating the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. When this program inputs the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of a physical quantity of a first composite material whose physical quantity value is unknown, The approximation function generation process is provided to generate an approximation function that outputs the value of .

The physical quantity estimating device in one embodiment is a physical quantity estimating device that becomes a component of a physical quantity estimating system that estimates the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. This physical quantity estimating device is based on the approximation function and the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of a physical quantity of a first composite material whose physical quantity value is unknown. A physical quantity estimator is provided that estimates a value of a physical quantity for the first composite material. Here, the approximation function is a function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable is input.

A program in one embodiment is a program that causes a computer to execute a process of estimating the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. This program is based on the approximation function and the value of the first categorical variable, which is a digital variable associated with a factor that affects the value of the physical quantity of the first composite material, the value of which is unknown. A physical quantity estimation process is provided to estimate the value of a physical quantity for a composite material. Here, the approximation function is a function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable is input.

The physical quantity estimation method in one embodiment is a physical quantity estimation method in which a computer estimates the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials. In this physical quantity estimation method, when inputting the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of a physical quantity of a first composite material whose physical quantity value is unknown, an approximation function generation step in which an approximation function generation section of the computer generates an approximation function that outputs a value of a physical quantity for the first composite material, and a physical quantity estimation section of the computer generates an approximation function based on the value of the first categorical variable and the approximation function; and a physical quantity estimating step of estimating the value of the physical quantity for.

According to one embodiment, the value of a physical quantity for a composite material can be estimated with high accuracy.

FIG. 2 is a diagram showing an example of a hardware configuration of a physical quantity estimation device. FIG. 2 is a functional block diagram showing the functions of the physical quantity estimating device in Embodiment 1. FIG. FIG. 2 is a diagram illustrating machine learning for generating an approximate function. 12 is a flowchart illustrating an operation for generating an approximation function. It is a flowchart explaining the operation|movement which estimates the value of the physical quantity with respect to the composite material of evaluation object. FIG. 2 is a functional block diagram showing an example in which a physical quantity estimation system is configured from a physical quantity estimation device and an approximation function generation device. 3 is a diagram showing a functional block configuration of a physical quantity estimating device in Embodiment 2. FIG. FIG. 2 is a diagram illustrating machine learning for generating an approximate function. 12 is a flowchart illustrating an operation for generating an approximation function. It is a flowchart explaining the operation|movement which estimates the value of the physical quantity with respect to the 1st composite material of evaluation object. It is a table showing data combining combination data and physical quantity data in a specific example. It is a table showing data in which combination-related data and additional data are combined with additional data in a specific example. It is a table showing first formulation data of a first composite material to be evaluated whose correspondence with physical quantity values is unknown. It is a table showing data including a first composite characteristic value, a first categorical variable, and a first irradiation amount in a specific example. It is a table showing estimation results of physical quantity values in a specific example. (a) is a graph showing the results of estimating the initial tensile strength of the first composite material using an approximation function that does not use categorical variables as input parameters, and (b) is a graph that shows the results of estimating the initial tensile strength of the first composite material using categorical variables as input parameters. It is a graph showing the result of estimating the initial tensile strength of the first composite material using an approximation function.

In all the drawings for explaining the embodiment, the same members are designated by the same reference numerals in principle, and repeated explanations thereof will be omitted. Note that, in order to make the drawings easier to understand, hatching may be added even in a plan view.

(Embodiment 1)
The technical concept of the first embodiment is a concept related to a physical quantity estimation system that estimates the value of a physical quantity corresponding to a blending ratio in a composite material in which a plurality of types of resins and compounding agents are combined.

Here, examples of the composite material include a wire coating material containing a resin and a compounding agent, and examples of the physical quantity include elongation and tensile strength of the composite material.

The resin is, for example, a polyolefin such as high density polyethylene, low density polyethylene, or ethylene acrylic acid copolymer, or an elastomer such as chlorinated polyethylene. On the other hand, examples of compounding agents include fillers such as talc, calcium carbonate, and silica, plasticizers, crosslinking agents, and stabilizers. However, the type and number of compositions such as resins and compounding agents constituting the composite material are not limited.

Note that the technical idea in the first embodiment is applicable not only to composite materials made of multiple types of resins and compounding agents, but also to composite materials made of multiple types of magnetic materials, and is applicable to physical quantities. For example, the magnetic susceptibility and the strength of the magnetic field (magnetic field, magnetic flux density) can be mentioned.

<Description of related technology>
First, related technology regarding a physical quantity estimation system that estimates the value of a physical quantity corresponding to a blending ratio will be described. The "related technology" used in this specification is a technology that is not a known technology, but has a problem discovered by the present inventor, and is a technology that is a premise of the present invention.

For example, in a physical quantity estimation system, if you input the material names and blending ratios of the constituent materials that make up a composite material, the physical quantity value of the composite material is calculated based on an approximation function that outputs the value of the physical quantity of the composite material. Related techniques for estimation can be considered. In this related technology, for example, data in which the name of the constituent materials, the blending ratio of the constituent materials, and the value of the physical quantity corresponding to this blending ratio are known is used as training data, the input is the material name and the blending ratio, and the output is the physical quantity. Generate an approximation function with the value of .

However, in the related technology, the object of estimating the value of the physical quantity is limited to the constituent materials included in the teacher data used to generate the approximate function. That is, if a constituent material that is not used to generate the approximate function is included in the composite material to be evaluated, the accuracy of estimating the value of the physical quantity of this composite material decreases. This is because, in the related technology, the input parameters of the approximation function are the names of the constituent materials, and therefore the input parameters cannot be synthesized. Let me explain this point clearly.

For example, data that associates a physical quantity value of ``100'' with the material name ``high-density polyethylene,'' and data that associates a physical quantity value of ``200'' with the material name ``low-density polyethylene.'' Suppose that we have generated an approximation function of related technology using as training data.

In this case, for example, the physical quantity values for a composite material that includes "high density polyethylene" and "low density polyethylene" as constituent materials and whose composition ratio is 50:50 are generated using related technology. Consider estimation using an approximation function.

First, in related technology, in order to obtain input parameters for a composite material, consider synthesizing input parameters for the constituent materials that make up the composite material. .5, and the calculation is "material name" x "numeric value", so the calculation itself of combining the input parameters does not make sense.

However, data that associates a physical quantity value of ``100'' with the material name ``high-density polyethylene'' and data that associates a physical quantity value of ``200'' with the material name ``low-density polyethylene'' is used as training data. Therefore, in this case, even without performing the calculation of combining the input parameters, the approximate function generated by the related technology allows the value of the physical quantity for the composite material to be "100" x 0.5 + "200" x 0. It is assumed that it can be estimated that 5=“150”. In other words, it is considered that with the related technology, it is possible to estimate the value of the physical quantity with high precision for a composite material containing "high density polyethylene" and "low density polyethylene" used for the training data.

On the other hand, for example, the physical quantity values for a composite material containing "polyolefin" and "high-density polyethylene" with a blending ratio of 70:30 as constituent materials of the composite material are generated using related technology. Consider estimation using an approximation function.

In this case as well, first of all, in order to obtain input parameters for the composite material, consider synthesizing the input parameters for the constituent materials that make up the composite material. Therefore, since the calculation is "material name" x "numeric value", the calculation itself of combining input parameters does not make any sense.

Furthermore, in this case, "polyolefin", which is not included in the teacher data, is included as a constituent material of the composite material. As a result, it is difficult to grasp the value of the physical quantity for "polyolefin" using the approximation function generated by the related technology, so the value of the physical quantity of the composite material is "???" x 0.7 + "100" ×0.3, making it difficult to estimate the value of the physical quantity with high precision for a composite material containing "polyolefin" and "high-density polyethylene."

This is due to the accuracy of estimating physical quantity values for composite materials that include constituent materials that are not used in the training data, because in an approximation function where the input parameter is the material name, it does not make sense to perform a composite calculation of the input parameters. decreases.

From the above, there is room for improvement in the related technology from the viewpoint of accurately estimating the value of a physical quantity for a composite material that includes constituent materials that are not used to generate the approximate function.

Therefore, in the first embodiment, improvements are made to address the room for improvement that exists in the related technology. In the following, the technical idea of the first embodiment in which this device is applied will be explained.

<Basic idea in Embodiment 1>
First, the inventors have found that the accuracy of estimating physical quantity values for composite materials that include constituent materials that are not used in the training data is as a result of using input parameters such as material names that are difficult to perform synthesis calculations on in related technologies. We focused on the fact that the essence of the problem lies in the decline in The inventor has also found that, for example, by using input parameters related to constituent materials that are easy to perform synthetic calculations on, it is possible to improve the estimation accuracy of physical quantity values for composite materials that include constituent materials that are not used in the training data. I have gained knowledge that this may be the case.

In this regard, the basic idea is that if you use parameters that can be expressed numerically as input parameters regarding constituent materials, it becomes possible to perform synthetic calculations. The idea is that the accuracy of estimating physical quantity values for composite materials can be improved. This point will be specifically explained below.

For example, the approximation function is created using data that associates the physical quantity value “100” with the input parameter “50” and data that associates the physical quantity value “150” with the input parameter “100” as training data. Suppose that it is generated.

Regarding this point, first, for example, a composite material that includes a constituent material with an input parameter of "50" and a constituent material with an input parameter of "100" and whose blending ratio of the constituent materials is 50:50. Consider estimating the value of a physical quantity for a material using the approximation function described above. In this case, considering that input parameters for the constituent materials that make up the composite material are synthesized in order to obtain input parameters for the composite material, "50" x 0.5 + "100" x 0.5 = "75". , the calculation is ``numeric value'' x ``numeric value'', so the calculation of combining input parameters can be easily performed. As a result, the input parameter "75" for the composite material can be obtained, and by inputting this input parameter "75" into the approximation function in the basic idea, it is possible to estimate the value of the physical quantity for the composite material. Therefore, according to the basic idea, it is possible to estimate the values of physical quantities with high accuracy for a composite material including the constituent materials used for the training data.

Next, for example, physical quantities for a composite material that includes a constituent material with an input parameter of "50" and a constituent material with an input parameter of "75" and whose blending ratio of the constituent materials is 50:50 are determined. Consider estimating the value of using the approximation function described above.

In this case, the constituent material of the composite material includes a constituent material of input parameter "75" which is not included in the teacher data. However, the basic idea is to use parameters expressed as numerical values as input parameters. Therefore, the basic idea is that input parameters for the constituent materials constituting the composite material can be synthesized in order to obtain input parameters for the composite material. Specifically, considering the synthesis of input parameters for the constituent materials that make up the composite material, "50" x 0.5 + "75" x 0.5 = "62.5", which means "numeric value" x "numeric value" '', it is possible to easily perform the operation of combining input parameters. As a result, the input parameter "62.5" for the composite material can be obtained, and by inputting this input parameter "62.5" into the approximation function in the basic idea, the value of the physical quantity for the composite material can be estimated. Can be done. Therefore, it can be seen that, according to the basic idea, the values of physical quantities can be estimated with high accuracy even for composite materials that include new constituent materials that are not used in the training data. This is basically a result of using input parameters that are easy to synthesize, parameters that can be expressed numerically as input parameters related to the constituent materials. As described above, the essence of the basic idea lies in the use of numerical values that can be subjected to synthetic calculations as input parameters regarding constituent materials.

Here, the present inventor focuses on the characteristic values of the constituent materials as parameters that can be expressed numerically as input parameters regarding the constituent materials.

That is, the basic idea of the first embodiment is to calculate the value of the physical quantity for the composite material by using an approximation function generated based on the characteristic values (numerical values) of the constituent materials constituting the composite material and the blending ratio of the constituent materials. The idea is to estimate the

According to this basic idea, the following effects can be obtained by using an approximation function generated based on the characteristic values of the constituent materials and the blending ratio of the constituent materials.

For example, in a related technology, an approximation function generated based on the names of constituent materials and the blending ratios of the constituent materials uses the input parameters as the names of the constituent materials and the blending ratios of the constituent materials. From this, if a new constituent material that was not used to generate the approximate function is included in the composite material to be evaluated, the material name of the constituent material used to generate the approximate function and the new Since the concept of a composite operation with the material name of the constituent material is meaningless, the accuracy of estimating the value of the physical quantity for this composite material will be reduced. That is, the approximation function generated by the related technology has a narrow application range for composite materials that can estimate the value of a physical quantity with high accuracy.

In particular, in the related technology, even if the name of the new constituent material (new material name) that was not used to generate the approximation function is known, if the value of the corresponding physical quantity is not known, the approximation generated by the related technology Using functions, it is not possible to estimate with high precision the physical quantities of the composite material to be evaluated. In other words, with the approximation function generated by the related technology, it is possible to estimate the physical quantity with high accuracy only for a composite material that includes only the constituent materials used in the training data.

On the other hand, the basic idea of the first embodiment is to generate an approximate function based on the characteristic values of the constituent materials and the blending ratio of the constituent materials. In this case, even if the composite material to be evaluated includes a new constituent material that was not used to generate the approximation function, if the property values corresponding to this new constituent material are known, the It is possible to estimate the values of physical quantities for materials with high accuracy. This is because the characteristic values of the constituent materials can be expressed numerically, making it possible to perform synthetic calculations.

In this way, the applicability range of the approximation function generated using the basic idea is wider than the applicability range of the approximation function generated using the related technology, and even if some of the composite materials to be evaluated are not used in the training data. This has great technical significance in that it is possible to estimate the physical quantities of composite materials with high accuracy even when new constituent materials are included. In other words, the applicability range of the approximation functions generated using related technology is limited to the range of the training data, but the applicability range of the approximation functions generated using the basic idea is not limited to the range of the training data, which is why the basic idea is superior. It can be said that it is a technical idea. For example, according to the basic idea, by accumulating property values for new constituent materials that were not used to generate the approximation function as a database, it is possible to Physical quantities can be estimated with high accuracy. Furthermore, by using the approximation function in the basic idea, even if it is a new constituent material that is not stored in the database, if the characteristic values of this new constituent material can be obtained by some means, it is possible to obtain a new constituent material. Approximate functions generated based on the basic idea have a wide range of applications in that they can estimate the physical quantities of composite materials with high precision.

Here, "characteristic values" refer to, for example, thermal properties, mechanical properties, physical properties, etc. For example, thermal properties include heat of fusion, melt flow rate, and the like. Further, the physical properties include specific gravity. On the other hand, "physical quantities" are assumed to include elongation, tensile strength, etc.

In this specification, "characteristic value" and "value of physical quantity" are used with a clear distinction. Specifically, the "characteristic value" is a parameter used to generate and input an approximation function. On the other hand, the "value of physical quantity" is a value output from the approximation function, and is a target value estimated by the physical quantity estimation system in the first embodiment.

In the following, we will mainly explain an example in which a physical quantity estimation system that embodies this basic idea is configured from a single computer, but the physical quantity estimation system in the first embodiment is a distributed system consisting of multiple computers. It is also possible to realize this.

<Configuration of physical quantity estimation device>
<<Hardware configuration>>
First, the hardware configuration of the physical quantity estimating device in the first embodiment will be explained.

FIG. 1 is a diagram showing an example of the hardware configuration of a physical quantity estimating device 100 according to the first embodiment. Note that the configuration shown in FIG. 1 merely shows an example of the hardware configuration of the physical quantity estimation device 100, and the hardware configuration of the physical quantity estimation device 100 is not limited to the configuration shown in FIG. It may be a configuration.

In FIG. 1, a physical quantity estimating device 100 includes a CPU (Central Processing Unit) 101 that executes a program. The CPU 101 is electrically connected to, for example, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and a hard disk device 112 via a bus 113, and controls these hardware devices. It is configured as follows.

Further, the CPU 101 is also connected to an input device and an output device via a bus 113. Examples of input devices include the keyboard 105, the mouse 106, the communication board 107, and the scanner 111. On the other hand, examples of output devices include the display 104, the communication board 107, and the printer 110. Further, the CPU 101 may be connected to, for example, a removable disk device 108 or a CD/DVD-ROM device 109.

The physical quantity estimating device 100 may be connected to a network, for example. For example, when the physical quantity estimating device 100 is connected to other external devices via a network, the communication board 107 that constitutes a part of the physical quantity estimating device 100 can be connected to a LAN (local area network), a WAN (wide area network), etc. or connected to the Internet.

The RAM 103 is an example of volatile memory, and the recording media of the ROM 102, removable disk device 108, CD/DVD-ROM device 109, and hard disk device 112 are examples of nonvolatile memory. These volatile memories and nonvolatile memories constitute a storage device of the physical quantity estimating device 100.

The hard disk device 112 stores, for example, an operating system (OS) 201, a program group 202, and a file group 203. The programs included in the program group 202 are executed by the CPU 101 while using the operating system 201. Further, the RAM 103 temporarily stores at least a portion of the operating system 201 programs and application programs to be executed by the CPU 101, and also stores various data necessary for processing by the CPU 101.

The ROM 102 stores a BIOS (Basic Input Output System) program, and the hard disk drive 112 stores a boot program. When the physical quantity estimating device 100 is started, the BIOS program stored in the ROM 102 and the boot program stored in the hard disk drive 112 are executed, and the operating system 201 is started by the BIOS program and the boot program.

The program group 202 stores programs that implement the functions of the physical quantity estimating device 100, and this program is read and executed by the CPU 101. Further, the file group 203 stores information, data, signal values, variable values, and parameters indicating the results of processing by the CPU 101 as each item of the file.

The file is recorded on a recording medium such as the hard disk device 112 or memory. Information, data, signal values, variable values, and parameters recorded on recording media such as the hard disk device 112 and memory are read by the CPU 101 to the main memory and cache memory, and are extracted, searched, referenced, compared, calculated, processed, and It is used for operations of the CPU 101, such as editing, outputting, printing, and displaying. For example, during the operation of the CPU 101 described above, information, data, signal values, variable values, and parameters are temporarily stored in main memory, registers, cache memory, buffer memory, and the like.

The functions of the physical quantity estimating device 100 may be realized by firmware stored in the ROM 102, or may be realized by only software, only hardware such as elements, devices, boards, and wiring, or a combination of software and hardware. , and may also be realized in combination with firmware. The firmware and software are recorded as programs on a recording medium such as a hard disk device 112, a removable disk, a CD-ROM, a DVD-ROM, and the like. The program is read and executed by the CPU 101. That is, the program causes the computer to function as the physical quantity estimating device 100.

In this way, the physical quantity estimation device 100 includes a CPU 101 as a processing device, a hard disk device 112 and memory as storage devices, a keyboard 105, a mouse 106, a communication board 107 as input devices, a display 104 as an output device, a printer 110, This is a computer equipped with a communication board 107. The functions of the physical quantity estimating device 100 are realized using a processing device, a storage device, an input device, and an output device.

<<Functional block configuration>>
Next, the functional block configuration of the physical quantity estimating device 100 will be explained.

FIG. 2 is a functional block diagram showing the functions of the physical quantity estimating device.

The physical quantity estimation device 100 includes an input section 301, a characteristic value data extraction section 302, a composite characteristic value calculation section 303, a composition related data generation section 304, an approximation function generation section 305, a physical quantity estimation section 306, and an output section. 307 and a data storage section 308.

The input unit 301 is configured to input characteristic value data. Here, "characteristic value data" refers to data that associates material names and material characteristic values for each of a plurality of different materials. The characteristic value data input to the input section 301 is stored in the data storage section 308. This data storage unit 308 functions as a database that stores a plurality of characteristic value data.

Further, the input unit 301 is configured to input formulation data and physical quantity data of a composite material that includes two or more materials contained in a plurality of different materials as constituent materials. Here, the "compounding data" is data including the material names and blending ratios of constituent materials constituting the composite material, and may also be referred to as blending information. On the other hand, "physical quantity data" is data indicating the value of a physical quantity in a composite material whose value of the physical quantity is known, and is, for example, data obtained by experiment. The formulation data and physical quantity data input to the input section 301 are also stored in the data storage section 308.

The characteristic value data extraction unit 302 is configured to extract characteristic value data corresponding to the constituent materials included in the composite material from among the plurality of characteristic value data stored in the data storage unit 308. For example, when the constituent materials included in the composite material are "polyolefin" and "polyethylene," the characteristic value data extraction unit 302 selects the characteristic value data corresponding to "polyolefin" and "polyethylene" from among the plurality of characteristic value data. The system is configured to extract characteristic value data corresponding to "Polyethylene".

The composite characteristic value calculation unit 303 configures a composite material based on the blending ratio included in the formulation data input to the input unit 301 and the characteristic values included in the characteristic value data extracted by the characteristic value data extraction unit 302. The composite material is configured to calculate composite characteristic values by performing calculations to synthesize characteristic values corresponding to constituent materials.

For example, consider a composite material in which the constituent materials include a constituent material with a characteristic value of "50" and a constituent material with a characteristic value of "75", and the blending ratio of the constituent materials is 50:50. In this case, the composite characteristic value calculation unit 303 calculates the composite characteristic value "62.5" by performing a composite operation of "50" x 0.5 + "75" x 0.5 = "62.5".

The synthetic characteristic values include, for example, the synthetic heat of fusion of the composite material, the synthetic melt flow rate of the composite material, and the like.

The synthesis-related data generation unit 304 is configured to generate synthesis-related data that associates the synthesis characteristic value calculated by the synthesis characteristic value calculation unit 303 with the value of the physical quantity for the composite material (“physical quantity data” of the composite material). has been done. This synthesis-related data generation is performed for the composite material input to the input unit 301 and for which the corresponding physical quantity is known. For example, if the value of the physical quantity for the composite material in the above example is "150", the synthesis-related data generation unit 304 generates a synthesis-related data that associates the synthesis characteristic value "62.5" with the physical quantity value "150". Generate data. The generated synthesis related data is stored in the data storage unit 308.

The approximation function generation unit 305 has a function of generating an approximation function based on the synthesis related data generated by the synthesis related data generation unit 304. That is, the approximation function generation unit 305 is configured to generate an approximation function that associates the composite characteristic value with the value of the physical quantity.

Specifically, as shown in FIG. 3, the approximation function generation unit 305 is configured to generate an approximation function whose input is the composite characteristic value and whose output is the value of the physical quantity, using the synthesis related data as training data. ing.

Here, the "approximation function" is defined as a function that, when a composite characteristic value is input, outputs a value of a physical quantity according to the composite characteristic value. In other words, an "approximation function" is a function that outputs the value of the physical quantity that is estimated to be realized by this composite material when the composite characteristic value of the composite material whose correspondence with the value of the physical quantity is unknown is input. is defined as In this way, the approximation function can be said to be a function used to estimate the value of a physical quantity for a composite material whose correspondence with the value of a physical quantity is unknown.

The physical quantity estimation unit 306 calculates the composite characteristic value calculation unit 303 based on the first blending ratio included in the first formulation data of the first composite material and the characteristic value of the first characteristic value data extracted by the characteristic value data extraction unit 302. The system is configured to estimate the value of the physical quantity corresponding to the first composite material based on the first composite characteristic value calculated in , and the approximation function generated by the approximation function generation unit 305 .

Note that the "first composite material" is a composite material that includes two or more materials contained in a plurality of different materials as constituent materials, and is a composite material for which the value of the corresponding physical quantity is unknown, and is the subject of evaluation. Represents a composite material. Here, the blending data of the first composite material is referred to as "first blending data", and the blending ratio included in the blending data of the first composite material is referred to as "first blending ratio". Further, the synthetic characteristic value of the first composite material is referred to as a "first synthetic characteristic value", and among the characteristic value data stored in the data storage unit 308, the characteristic value corresponding to the constituent material included in the first composite material The data is called "first characteristic value data."

For example, the first combination data is input to the physical quantity estimating device 100 from the input section 301, and the first characteristic value data is extracted by the characteristic value data extraction section 302.

The output unit 307 outputs the value of the physical quantity estimated by the physical quantity estimation unit 306.

In this way, the physical quantity estimating device 100 is configured.

The constituent materials of the composite material include, for example, a plurality of different types of resins, but other constituent materials may also be included. For example, the constituent materials of the composite material may include additives, antioxidants, crosslinking aids, and the like. Moreover, a crosslinked resin can be mentioned as an example of the resin. Since additional functions are added to the physical quantity estimating device 100 depending on the specific constituent materials of the composite material, this point will be explained below.

<<<When the composite material contains additives>>>
When an additive is included in the composite material, in addition to the above-mentioned functions, the synthetic characteristic value calculation unit 303 further calculates the average inter-filler distance of the additive or the volume fraction of the additive based on the characteristic value of the additive. is configured to calculate. The composite characteristic value calculated by the composite characteristic value calculation unit 303 includes the average inter-filler distance of the additive or the volume fraction of the additive.

Note that the average interparticle distance represented by the average interfiller distance is calculated, for example, from the average particle diameter D50 using a theoretical formula. Further, the volume fraction is calculated from the specific gravity of the compounded materials.

<<<When the composite material contains an antioxidant and a crosslinking aid>>>
When the composite material contains an antioxidant and a crosslinking aid, in addition to the above-mentioned functions, the synthesis characteristic value calculation unit 303 further calculates the antioxidant value based on the characteristic value of the antioxidant and the characteristic value of the crosslinking aid. The system is configured to calculate the number of reactive moles of the primary reactive group of the agent, the reactive mole number of the secondary reactive group of the antioxidant, and the reactive mole number of the crosslinking aid. The synthetic characteristic values calculated by the synthetic characteristic value calculation unit 303 include the number of reactive moles of the primary reactive group of the antioxidant, the reactive mole number of the secondary reactive group of the antioxidant, and the reactive mole number of the crosslinking aid. is included.

<<<When the composite material contains crosslinked resin>>>
When the composite material includes a crosslinked resin, the input unit 301 is configured to further input the radiation dose for crosslinking the resin. The approximation function generation unit 305 is configured to generate an approximation function based on the synthesis related data and the radiation dose. The approximation function in this case is generated as a function whose inputs are the composite characteristic value and the radiation dose, and whose output is the value of the physical quantity. Further, the physical quantity estimating unit 306 is configured to estimate the value of the physical quantity for the first composite material based on the first composite characteristic value, the first radiation dose, and the approximation function. Here, the "first radiation dose" represents the radiation dose irradiated to the first composite material.

<Operation of physical quantity estimation device>
The physical quantity estimating device 100 according to the first embodiment is configured as described above, and its operation will be described below. The operations of the physical quantity estimating device 100 include "an operation of generating an approximation function" and "an operation of estimating the value of a physical quantity corresponding to a composite material to be evaluated." Therefore, these operations will be explained below.

<<Generation operation of approximate function>>
FIG. 4 is a flowchart illustrating the approximate function generation operation.

In FIG. 4, first, the input unit 301 inputs a plurality of characteristic value data in which material names and material characteristic values are associated with each other for each of a plurality of different materials (S101). Then, the plurality of characteristic value data input to the input section 301 are stored in the data storage section 308 (S102).

Next, the input unit 301 inputs formulation data and physical quantity data of a composite material that includes two or more materials as constituent materials and whose corresponding physical quantity values are known (S103).

Thereafter, the characteristic value data extraction unit 302 extracts characteristic value data corresponding to the constituent materials included in the composite material from among the plurality of characteristic value data stored in the data storage unit 308 (S104). Next, the composite characteristic value calculation unit 303 calculates the composite material by performing an operation to synthesize the characteristic values of the characteristic value data extracted by the characteristic value data extraction unit based on the formulation data input from the input unit 301. A composite characteristic value is calculated (S105).

Then, the synthesis-related data generation unit 304 generates synthesis-related data that associates the synthesis characteristic value calculated by the synthesis characteristic value calculation unit 303 with the value of the physical quantity (physical quantity data) for the composite material (S106). Thereafter, the synthesis related data generated by the synthesis related data generation unit 304 is stored in the data storage unit 308 (S107).

Next, the approximation function generation unit 305 generates an approximation function based on the synthesis related data generated by the synthesis related data generation unit 304 (S108). Specifically, the approximation function generation unit 305 uses the synthesis related data as teacher data to generate an approximation function whose input is the synthesis characteristic value and whose output is the value of the physical quantity (see FIG. 3).

The approximate function generated by the approximate function generator 305 is then stored in the data storage unit 308 (S109). In this way, the approximate function generation operation is performed.

<<Operation for estimating the value of physical quantity for the first composite material to be evaluated>>
Next, the operation of estimating the value of the physical quantity for the first composite material to be evaluated will be explained.

FIG. 5 is a flowchart illustrating the operation of estimating the value of the physical quantity for the first composite material to be evaluated. Note that the approximation function has already been stored in the data storage unit 308.

In FIG. 5, first, the input unit 301 inputs first formulation data of a first composite material to be evaluated whose correspondence with the value of a physical quantity is unknown (S201).

Next, the characteristic value data extraction unit 302 extracts first characteristic value data corresponding to the constituent materials included in the first composite material from among the plurality of characteristic value data stored in the data storage unit 308. (S202).

Subsequently, the composite characteristic value calculation unit 303 performs an operation to synthesize the characteristic values of the first characteristic value data extracted by the characteristic value data extraction unit 302 based on the first combination data input from the input unit 301. Accordingly, the first composite characteristic value of the first composite material is calculated (S203).

After that, the physical quantity estimation unit 306 estimates the value of the physical quantity for the first composite material by inputting the first composite characteristic value calculated by the composite characteristic value calculation unit 303 into the approximation function (S204). Then, the output unit 307 outputs the value of the physical quantity estimated by the physical quantity estimation unit 306 (S205). In this manner, the physical quantity estimating device 100 can output a physical quantity value that is likely to be realized for the first composite material to be evaluated whose correspondence with the physical quantity value is unknown.

<Physical quantity estimation program>
The physical quantity estimation method implemented by the physical quantity estimation device 100 described above can be realized by a physical quantity estimation program that causes a computer to execute the physical quantity estimation process.

For example, in the physical quantity estimating device 100 consisting of a computer shown in FIG. 1, the physical quantity estimating program according to the first embodiment can be introduced as one of the program group 202 stored in the hard disk drive 112. Then, by causing the computer serving as the physical quantity estimation device 100 to execute this physical quantity estimation program, the physical quantity estimation method according to the first embodiment can be realized.

A physical quantity estimation program that causes a computer to execute each process for creating data related to the physical quantity estimation process can be recorded on a computer-readable recording medium and distributed. Examples of recording media include magnetic storage media such as hard disks and flexible disks, optical storage media such as CD-ROMs and DVD-ROMs, and hardware devices such as non-volatile memories such as ROMs and EEPROMs. is included.

<Modified example>
In Embodiment 1, as shown in FIG. 2, an example has been described in which a physical quantity estimation system for estimating the value of a physical quantity for a composite material is configured from a single physical quantity estimation device 100. However, the present invention is not limited to this, and it can also be constructed from a distributed system, for example.

FIG. 6 is a functional block diagram showing an example in which a physical quantity estimation system is configured from a physical quantity estimation device and an approximation function generation device.

As shown in FIG. 6, the physical quantity estimation system includes a physical quantity estimation device 400 and an approximate function generation device 500. For example, the physical quantity estimation device 400 and the approximate function generation device 500 are connected through a network 600.

The physical quantity estimation device 400 includes an input section 301A, a first characteristic value data extraction section 302A, a first composite characteristic value calculation section 303A, a physical quantity estimation section 306, an output section 307, a communication section 309A, and a data storage section. It has 310A.

The approximation function generation device 500 includes an input section 301B, a characteristic value data extraction section 302B, a composite characteristic value calculation section 303B, a synthesis related data generation section 304, an approximation function generation section 305, a communication section 309B, and a data storage. It has a section 310B.

The physical quantity estimating device 400 and the approximate function generating device 500 configured as described above are configured to be able to transmit and receive data through the communication unit 309A and the communication unit 309B via the network 600. Then, in the approximation function generation device 500, the above-described “approximation function generation operation” is performed to generate an approximation function.

On the other hand, the physical quantity estimating device 400 outputs the first composite characteristic value calculated by the first composite characteristic value calculation unit 303A to the approximate function generating device 500. Thereafter, in the approximation function generation device 500, the output result output from the approximation function by inputting the first composite characteristic value to the approximation function is input from the approximation function generation device 500, and is stored in the data storage unit 310A.

Thereafter, the physical quantity estimating device 400 acquires the value of the physical quantity for the composite material to be evaluated, based on the output result input from the approximate function generating device 500.

In this way, the physical quantity estimation system according to the first embodiment can also be constructed by a distributed system including the physical quantity estimation device 400 and the approximate function generation device 500.

(Embodiment 2)
In the first embodiment described above, in the first composite material in which the value of the physical quantity is unknown, based on the first blending ratio of the first composite material and the characteristic value corresponding to the constituent materials included in the first composite material, By calculating the first composite characteristic value of the first composite material and inputting the calculated first composite characteristic value into the approximation function generated by the approximation function generation section, the output value output from the approximation function is converted into the first composite material. It is estimated as the value of physical quantity for the material.

In this case, even if the first composite material includes a new constituent material that was not used in the teacher data, the value of the physical quantity for the first composite material can be estimated with high accuracy.

In other words, even if the first composite material includes a new constituent material that was not used to generate the approximation function, if the characteristic values corresponding to this new constituent material are known, the physical quantities for the first composite material can be calculated. The value of can be estimated with high accuracy. This is because the characteristic values of the constituent materials are expressed numerically, which makes it possible to perform synthetic calculations and calculate the first composite characteristic value (analog numerical value) for the first composite material. be.

Therefore, the technical idea in the first embodiment is that even if the first composite material to be evaluated includes a new constituent material that was not used in the training data, the physical quantity for the first composite material is It has great technical significance in that it can estimate values with high precision.

As described above, the technical idea in the first embodiment is to focus on "characteristic values" that can be synthesized as factors that influence the values of physical quantities of the first composite material, and to calculate the synthetic properties of the first composite material. By using the value as an input parameter (explanatory variable) of the approximation function, the output value from the approximation function is used as an estimated value of the physical quantity of the first composite material.

In this regard, it is considered that the factors that influence the values of the physical quantities of the first composite material are not limited to the "characteristic values" that can be subjected to the above-mentioned synthesis calculation. That is, input parameters other than the "characteristic values" can be considered as factors that influence the values of the physical quantities of the first composite material.

Therefore, in the second embodiment, a technical concept for estimating the value of the physical quantity for the first composite material with high accuracy by considering input parameters other than the "characteristic value" will be described.

<Basic idea in Embodiment 2>
The basic idea of the second embodiment is that when the value of a categorical variable consisting of a digital variable associated with a factor that affects the value of a physical quantity of a first composite material whose physical quantity value is unknown is input, The idea is to estimate the value of the physical quantity of the first composite material by using an approximation function that outputs the value of the physical quantity for the composite material. That is, the basic idea of the second embodiment is to use digital numerical values such as the values of categorical variables instead of using analog numerical values such as "characteristic values" as input parameters of the approximation function. be.

In other words, the basic idea of the second embodiment is that when inputting the value of a categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of the first composite material whose physical quantity value is unknown, The idea is to generate an approximation function that outputs the value of the physical quantity for the first composite material, and estimate the value of the physical quantity for the first composite material based on the value of the categorical variable and the approximation function.

As a result, for example, factors that influence the value of the physical quantity of the first composite material include not only factors suitable for being expressed as analog numerical values such as "characteristic values" but also factors such as the values of categorical variables. Factors suitable for being expressed in digital numbers can also be considered.

Below, an example of a factor suitable for being expressed as a digital numerical value such as a value of a categorical variable among the factors that influence the value of the physical quantity of the first composite material will be explained.

For example, consider a material containing a resin and a flame retardant as the first composite material. As physical quantities of the first composite material, elongation and tensile strength are considered. Here, in order to improve elongation and tensile strength, surface treatment is performed on the surface of the flame retardant mixed with the resin. This surface treatment includes silane coupling treatment and fatty acid treatment. Furthermore, there are a plurality of types of materials constituting flame retardants, typified by magnesium hydroxide and aluminum hydroxide.

Therefore, it is thought that the elongation and tensile strength of the first composite material are influenced by the type of surface treatment performed on the surface of the flame retardant and the type of material that constitutes the flame retardant. In other words, the type of surface treatment performed on the surface of the flame retardant and the type of material constituting the flame retardant can be considered factors that affect the value of the physical quantity of the first composite material.

Therefore, by using the type of surface treatment performed on the surface of the flame retardant and the type of material constituting the flame retardant as input parameters of the approximation function, it is possible to influence the value of the physical quantity of the first composite material. Since it is possible to incorporate the influencing factors into the approximation function, it is thought that it is possible to generate an approximation function that can output highly accurate estimated values of physical quantities.

Regarding this point, when using the surface treatment performed on the surface of the flame retardant or the type of material constituting the flame retardant as an input parameter of the approximation function, a categorical variable consisting of digital variables should be adopted as the input parameter. is suitable. This is because, for example, a categorical variable that is a digital variable consists of a first variable indicating the presence or absence of the first surface treatment of the flame retardant, a second variable indicating the presence or absence of the second surface treatment of the flame retardant, and the flame retardant. This is because it can be configured to include a third variable indicating the presence or absence of the first material and a fourth variable indicating the presence or absence of the second material constituting the flame retardant. The first surface treatment is defined as silane coupling treatment, while the second surface treatment is defined as fatty acid treatment, the first material is magnesium hydroxide, and the second material is hydroxide. By defining aluminum as categorical variables, the type of surface treatment performed on the surface of the flame retardant and the type of material constituting the flame retardant can be used as input parameters of the approximation function.

Specifically, when the value of the first variable, which is a digital variable, is "1", it indicates that silane coupling treatment has been carried out as a surface treatment for the flame retardant, while the value of the first variable, which is a digital variable, is "1". In the case of "0", it can be expressed that silane coupling treatment is not performed as surface treatment of the flame retardant. Similarly, when the value of the second variable, which is a digital variable, is "1", it indicates that fatty acid treatment is carried out as a surface treatment for the flame retardant, while the value of the second variable, which is a digital variable, is "0". In the case of , it can be indicated that fatty acid treatment has not been carried out as a surface treatment for the flame retardant. In addition, when the value of the third digital variable is "1", it indicates that the material of the flame retardant is magnesium hydroxide, while when the value of the third digital variable is "0", it indicates that the flame retardant material is magnesium hydroxide. This can indicate that the material of the fuel is not magnesium hydroxide. Similarly, when the value of the fourth variable, which is a digital variable, is "1", it indicates that the material of the flame retardant is aluminum hydroxide, while when the value of the fourth variable, which is a digital variable, is "0", It can indicate that the flame retardant material is not aluminum hydroxide.

From the above, for example, the type of surface treatment performed on the surface of the flame retardant and the type of material constituting the flame retardant are among the factors that influence the physical quantity values of the first composite material. It can be said that it is a factor suitable for being expressed as a digital value like the value of a variable. Therefore, by using the type of surface treatment performed on the surface of the flame retardant and the type of material constituting the flame retardant as input parameters of the approximation function, an approximation that can output highly accurate estimated values of physical quantities is possible. The basic idea in the second embodiment of using the value of a categorical variable consisting of a digital variable as an input parameter is useful in order to generate a function.

<Embodiment mode that embodies the basic idea>
Next, an implementation mode that embodies the basic idea of the second embodiment will be described. For example, the physical quantity estimating device 100 can also be used in an embodiment that embodies the basic idea of the second embodiment.

<<Configuration of physical quantity estimation device>>
<<<Hardware configuration>>>
The hardware configuration of the physical quantity estimating device 100 in the second embodiment is, for example, similar to the hardware configuration shown in FIG. 1. Note that the configuration shown in FIG. 1 merely shows an example of the hardware configuration of the physical quantity estimation device 100, and the hardware configuration of the physical quantity estimation device 100 is not limited to the configuration shown in FIG. It may be a configuration.

<<<Functional block configuration>>>
FIG. 7 is a diagram showing a functional block configuration of the physical quantity estimating device 100 in the second embodiment. In FIG. 7, the physical quantity estimation device 100 includes an input section 301, an additional data generation section 304A, an approximation function generation section 305, a physical quantity estimation section 306, an output section 307, and a data storage section 308.

The input unit 301 is configured to input the value of a categorical variable, which is a digital variable associated with a factor that affects the value of the physical quantity of the composite material. For example, the categorical variables include a first variable that indicates the presence or absence of silane coupling treatment of the flame retardant, a second variable that indicates the presence or absence of fatty acid treatment of the flame retardant, and a second variable that indicates the presence or absence of magnesium hydroxide constituting the flame retardant. 3 variables and a fourth variable indicating the presence or absence of aluminum hydroxide constituting the flame retardant, the input section 301 contains the value of the first variable, the value of the second variable, and the fourth variable indicating the presence or absence of aluminum hydroxide constituting the flame retardant. The values of the three variables and the value of the fourth variable are input. The value of the categorical variable input to the input unit 301 is stored in the data storage unit 308. This data storage unit 308 functions as a database that stores the values of a plurality of categorical variables.

Further, the input unit 301 is configured to input formulation data and physical quantity data of a composite material that includes two or more materials contained in a plurality of different materials as constituent materials.

Here, the "composition data" is data including the material names and blending ratios of constituent materials constituting the composite material, and is data called blending information. On the other hand, "physical quantity data" is data indicating the value of a physical quantity in a composite material whose value of the physical quantity is known, and is, for example, data obtained by experiment. The formulation data and physical quantity data input to the input section 301 are also stored in the data storage section 308.

The additional data generation unit 304A generates additional data that associates the value of the categorical variable input to the input unit 301 and stored in the data storage unit 308 with the value of the physical quantity for the composite material (“physical quantity data” of the composite material). is configured to generate.

This generation of additional data is performed for the composite material input to the input unit 301 and for which the corresponding physical quantity is known.

The approximation function generation unit 305 has a function of generating an approximation function based on the values of the categorical variables input to the input unit 301 and stored in the data storage unit 308. In other words, the approximation function generation unit 305 has a function of generating an approximation function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable of the first composite material whose value of the physical quantity is unknown is input. has. That is, the approximation function generation unit 305 is configured to generate an approximation function that associates the value of the categorical variable with the value of the physical quantity.

Specifically, as shown in FIG. 8, the approximation function generation unit 305 is configured to generate an approximation function whose input is the value of a categorical variable and whose output is the value of a physical quantity, using additional data as training data. has been done. Here, the "approximation function" is defined as a function that, when a value of a categorical variable is input, outputs a value of a physical quantity according to the value of this categorical variable. In other words, the "approximation function" means that when the value of the first categorical variable of the constituent material of the first composite material whose correspondence with the value of the physical quantity is unknown is input, the "approximation function" is realized in the first composite material. It is defined as a function that outputs the value of a physical quantity estimated to be In this way, the approximation function can be said to be a function used to estimate the value of the physical quantity for the first composite material whose correspondence with the value of the physical quantity is unknown.

The physical quantity estimation unit 306 estimates the value of the physical quantity corresponding to the first composite material based on the value of the first categorical variable of the first composite material and the approximation function generated by the approximation function generation unit 305. It is composed of Note that the "first composite material" is a composite material that includes two or more materials contained in a plurality of different materials as constituent materials, and is a composite material for which the value of the corresponding physical quantity is unknown, and is the subject of evaluation. Represents a composite material. Here, the categorical variable of the first composite material is referred to as a "first categorical variable."

The output unit 307 outputs the value of the physical quantity estimated by the physical quantity estimation unit 306.

In this way, the physical quantity estimating device 100 is configured.

<<Operation of physical quantity estimation device>>
<<<Generation operation of approximate function>>>
FIG. 9 is a flowchart illustrating the approximate function generation operation.

In FIG. 9, first, the input unit 301 inputs the value of a categorical variable, which is a digital variable associated with a factor that affects the value of the physical quantity of the composite material (S301). The value of the categorical variable is then stored in the data storage unit 308 (S302).

Next, the input unit 301 inputs formulation data and physical quantity data of a composite material that includes two or more materials as constituent materials and whose corresponding physical quantity values are known (S303).

Then, the additional data generation unit 304A generates additional data that associates the value of the categorical variable of the composite material with the value of the physical quantity (physical quantity data) for the composite material (S304). Thereafter, the additional data generated by the additional data generation unit 304A is stored in the data storage unit 308 (S305).

Subsequently, the approximate function generation unit 305 generates an approximate function based on the additional data generated by the additional data generation unit 304A (S306). Specifically, the approximation function generation unit 305 uses the additional data as teacher data to generate an approximation function whose input is the value of a categorical variable and whose output is the value of the physical quantity. Then, the approximation function generated by the approximation function generation unit 305 is stored in the data storage unit 308 (S307).

In this way, the approximate function generation operation is performed.

<<<Operation for estimating the value of physical quantity for the first composite material to be evaluated>>>
Next, the operation of estimating the value of the physical quantity for the first composite material to be evaluated will be explained.

FIG. 10 is a flowchart illustrating the operation of estimating the value of the physical quantity for the first composite material to be evaluated. Note that the approximation function has already been stored in the data storage unit 308.

In FIG. 10, first, the input unit 301 inputs the value of the first categorical variable of the first composite material to be evaluated whose correspondence with the value of the physical quantity is unknown (S401).

Next, the physical quantity estimation unit 306 estimates the value of the physical quantity for the first composite material by inputting the value of the first categorical variable into the approximation function (S402). Then, the output unit 307 outputs the value of the physical quantity estimated by the physical quantity estimation unit 306 (S403).

In this manner, the physical quantity estimating device 100 can output a value of a physical quantity that is likely to be realized for the first composite material to be evaluated whose correspondence with the value of the physical quantity is unknown.

<Specific example>
Next, a specific example in which Embodiment 1 and Embodiment 2 are combined will be described.

In a specific example, categorical variables, synthetic characteristic values, formulation information, and process condition values are used as input parameters.

FIG. 11 is a table showing data that is a combination of formulation data and physical quantity data in this specific example. In FIG. 11, the formulation number is an ID number that identifies the composite material. In FIG. 11, the constituent materials constituting the composite material include a resin, a flame retardant (filler), an antioxidant, a lubricant, a coloring agent, and a crosslinking aid.

As the resin, resins whose material names are Resin A, Resin B, Resin C, Resin D, Resin E, Resin F, and Resin G are listed. Further, as flame retardants, flame retardants whose material names are Flame Retardant H, Flame Retardant I, and Flame Retardant J are listed. Furthermore, examples of antioxidants include antioxidants with the material names Antioxidant K and Antioxidant L, and examples of lubricants include lubricants with material names Lubricant M, Lubricant N, and Lubricant O. . Further, examples of the coloring agent include a coloring agent whose material name is colorant P, and examples of the crosslinking aid include a crosslinking aid whose material name is crosslinking aid Q.

In FIG. 11, for example, the composite material specified by the formulation number "ID1" includes resin B (20 parts by mass), resin C (50 parts by mass), resin E (30 parts by mass), and flame retardant H. (200 parts by mass), antioxidant K (1 part by mass), antioxidant L (2 parts by mass), lubricant M (1 part by mass), lubricant N (2 parts by mass), colorant P (2 parts by mass) and crosslinking aid Q (4 parts by mass). In this manner, it can be seen that the data shown in FIG. 11 includes formulation data (composition information) including the material names and blending ratios of the constituent materials constituting the composite material.

Furthermore, in FIG. 11, for example, it is shown that the composite material specified by the formulation number "ID1" has a tensile strength as a physical quantity of "8.3347". That is, the data shown in FIG. 11 also includes physical quantity data indicating values of physical quantities in a composite material whose physical quantity values are known. From the above, it can be seen that FIG. 11 shows a combination of formulation data and physical quantity data for a composite material whose physical quantity values are known.

Next, in this specific example, data shown in FIG. 12 is generated based on the data shown in FIG. 11, which includes formulation data and physical quantity data. The data shown in FIG. 12 is composed of synthesis-related data, additional data, and additional data. Here, the synthesis-related data is data that associates synthesis characteristic values with physical quantity values (physical quantity data) for the composite material. Additionally, additional data is data that relates values of categorical variables and physical quantity values for composite materials (physical quantity data), and additional data refers to irradiation doses as process condition values and physical quantity values for composite materials. (physical quantity data).

First, the synthesis related data included in the data shown in FIG. 12 will be explained.

The synthetic characteristic values included in the synthesis-related data are obtained by performing an operation to synthesize the characteristic values of the characteristic value data based on the compounding ratio included in the compounding data in the data shown in FIG. 11 and characteristic value data (not shown). Calculated. Here, the characteristic value data refers to data that associates a material name with a material characteristic value for each of a plurality of different materials, and this characteristic value data is assumed to have been acquired in advance. For example, when focusing on resin A shown in FIG. 11, the characteristic value data can be said to be data that associates the material name of resin A with the characteristic value of resin A.

Here, the synthetic property values do not necessarily need to be calculated using the property values for all the constituent materials that make up the composite material, but only for some of the constituent materials that make up the composite material (for related constituent materials). The composition calculation may be performed using only the characteristic values for (limitation).

The composite characteristic value will be specifically explained below.

In FIG. 12, the composite characteristic values of this specific example include five types of composite characteristic values. Below, these five types of composite characteristic values will be explained.

(1) Filler volume ratio The filler volume ratio indicates the ratio of the volume of the filler (flame retardant) to the volume of the resin (base polymer), and represents the proportion of the filler added to the resin. It is a parameter. For example, the synthetic characteristic value regarding filler volume ratio relates to resin and filler. From this, among the constituent materials constituting the composite material, the respective characteristic values of the resins (resin A to resin G) shown in FIG. 11 and the flame retardants (flame retardant H to flame retardant J) shown in FIG. An operation is performed to calculate a composite characteristic value based on the respective characteristic values and the blending ratio shown in FIG. 11.

(2) Maleic anhydride modification amount The maleic anhydride modification amount is a parameter representing the amount of maleic anhydride (MAH) contained in the composite material. Maleic anhydride has the function of adhering resin and filler, so the amount of maleic anhydride is thought to affect the elongation and tensile strength of the resin composition, so it is used as a parameter. . For example, the synthetic property value regarding the amount of maleic anhydride modification is related to the resin. From this, synthetic characteristic values are calculated based on the respective characteristic values of the resins (Resin A to Resin G) shown in Fig. 11 and the blending ratios shown in Fig. 11 among the constituent materials constituting the composite material. Calculations are performed to do this.

(3) Amount of crystals The amount of crystals is a parameter representing the amount of crystalline resin contained in the composite material. Since the hardness of the composite material changes depending on the amount of crystalline resin, it is considered that the amount of crystalline resin affects the elongation and tensile strength of the resin composition, so it is used as a parameter. For example, the synthetic characteristic value regarding the amount of crystals is related to the resin. From this, synthetic characteristic values are calculated based on the respective characteristic values of the resins (Resin A to Resin G) shown in Fig. 11 and the blending ratios shown in Fig. 11 among the constituent materials constituting the composite material. Calculations are performed to do this.

(4) Vinyl acetate group weight The vinyl acetate group weight is a parameter representing the amount of vinyl acetate groups contained in the composite material. Since the hardness of the composite material changes depending on the amount of vinyl acetate groups, the amount of vinyl acetate groups is thought to affect the elongation and tensile strength of the resin composition, and is therefore used as a parameter. For example, the synthetic characteristic value regarding the amount of crystals is related to the resin. From this, synthetic characteristic values are calculated based on the respective characteristic values of the resins (Resin A to Resin G) shown in Fig. 11 and the blending ratios shown in Fig. 11 among the constituent materials constituting the composite material. Calculations are performed to do this.

(5) Filler surface area Filler surface area is used as a parameter representing the particle size of filler used as a flame retardant or flame retardant aid. The filler particle size is considered to affect the elongation and tensile strength of the resin composition, and is therefore used as a parameter. For example, synthetic properties related to filler surface area are related to flame retardants. From this, among the constituent materials constituting the composite material, based on the respective characteristic values of the flame retardants (flame retardant H to flame retardant J) shown in FIG. 11 and the blending ratio shown in FIG. An operation is performed to calculate the value.

Next, additional data included in the data shown in FIG. 12 will be explained.

The additional data includes categorical variables. The value of this categorical variable is input based on the formulation data included in the data shown in FIG.

The categorical variable is a digital variable, and for example, four types of categorical variables are used in the additional data. Specifically, the categorical variable "CAT1" indicates the type of surface treatment of the flame retardant, and when the variable value is "1", it indicates that the surface treatment of the flame retardant is silane coupling treatment. On the other hand, when the variable value is "0", it means that the silane coupling treatment is not performed.

The categorical variable "CAT2" indicates the type of surface treatment of the flame retardant, and when the variable value is "1", it means that the surface treatment of the flame retardant is fatty acid treatment. A value of "0" indicates that no fatty acid treatment has been performed.

The categorical variable "CAT3" indicates the component of the flame retardant, and if the variable value is "1", it means that the component of the flame retardant contains magnesium hydroxide, while if the variable value is "0" In the case of , it means that the flame retardant component does not contain magnesium hydroxide.

The categorical variable "CAT4" indicates the component of the flame retardant, and if the variable value is "1", it means that the component of the flame retardant contains aluminum hydroxide, while if the variable value is "0" In the case of , it means that the flame retardant component does not contain aluminum hydroxide.

For example, in FIG. 12, the composite material identified by the formulation number "ID1" has a categorical variable "CAT1" of "1", a categorical variable "CAT2" of "0", and a categorical variable "CAT3" is "1" and the categorical variable "CAT4" is "0". From this, it can be concluded that the composite material identified by the formulation number "ID1" contains magnesium hydroxide as a constituent flame retardant component, and has been subjected to silane coupling treatment as a surface treatment for the flame retardant. Recognize.

Furthermore, additional data included in the data shown in FIG. 12 will be explained.

The additional data includes the irradiation dose as a process condition value. This irradiation amount represents the amount of radiation irradiated in the step of crosslinking the resin contained in the composite material. For example, when the irradiation amount is "0", it means that the step of crosslinking the resin by radiation irradiation is not performed in the first place.

The data shown in FIG. 12 is configured as described above.

Next, an approximate function is generated based on the combination data included in FIG. 11 and the data shown in FIG. 12 (synthesis-related data, additional data, and additional data). Specifically, the formulation data included in FIG. 11 and the data shown in FIG. An approximate function is generated by performing machine learning. In this way, in this specific example, the approximation function is generated using the data shown in FIG. 12 as well as the recipe information including material names and compounding ratios (the recipe data included in FIG. 11) as the teacher data. In this case, an approximation function is generated whose inputs are formulation information, composite characteristic values, values of categorical variables, and values of irradiance, and whose outputs are physical quantity values.

Note that when generating the approximation function, not only are all of the formulation information, composite characteristic values, categorical variable values, and irradiance values input, but also the formulation information, composite characteristic values, and categorical variable values are input. Alternatively, the approximate function may be generated by performing machine learning using the value of the physical quantity as an output. In this case, an approximation function is generated whose inputs are combination information, composite characteristic values, and variable values of categorical variables, and whose outputs are physical quantity values. Alternatively, the approximate function may be generated by performing machine learning in which the values of categorical variables and process condition values (values of irradiation amount) are input, and the values of physical quantities are output. In this case, an approximation function is generated whose inputs are the values of categorical variables and process condition values, and whose output is the value of the physical quantity.

Furthermore, the approximate function may be generated by performing machine learning using the values of categorical variables as input and the values of physical quantities as output. In this case, an approximation function is generated whose input is a value of a categorical variable and whose output is a value of a physical quantity.

Next, a description will be given of estimating the value of the physical quantity for the first composite material to be evaluated based on the approximation function generated as described above.

FIG. 13 is a table showing first formulation data of a first composite material to be evaluated whose correspondence with physical quantity values is unknown. In FIG. 13, the formulation number is an ID number that identifies the first composite material. In FIG. 13, the constituent materials constituting the first composite material include a resin, a flame retardant (filler), an antioxidant, a lubricant, a coloring agent, and a crosslinking aid.

As the resin, resins whose material names are Resin A, Resin D, Resin F, and Resin G are listed. Further, as the flame retardant, a flame retardant whose material name is Flame Retardant I is listed. Further, examples of the antioxidant include antioxidants whose material names are Antioxidant K and Antioxidant L, and examples of lubricants include lubricants whose material names are Lubricant M and Lubricant O. Further, examples of the coloring agent include a coloring agent whose material name is colorant P, and examples of the crosslinking aid include a crosslinking aid whose material name is crosslinking aid Q.

In FIG. 13, for example, the first composite material specified by the formulation number "ID100" includes resin A (45 parts by mass), resin D (40 parts by mass), resin F (15 parts by mass), Fuel agent I (160 parts by mass), antioxidant K (1 part by mass), antioxidant L (2 parts by mass), lubricant M (1 part by mass), lubricant O (2 parts by mass), colorant P (2 parts by mass) part) and crosslinking aid Q (4 parts by mass). In this way, it can be seen that the data shown in FIG. 13 is first formulation data (mixture information) that includes the material names and compounding ratios of the constituent materials constituting the first composite material.

Next, in this specific example, data shown in FIG. 14 is generated based on the first formulation data shown in FIG. 13. The data shown in FIG. 14 is composed of the first composite characteristic value, the value of the first categorical variable, and the value of the first dose.

In this specific example, the first composite characteristic value of the first composite material is calculated based on the first formulation data shown in FIG. 13 and the first characteristic value data not shown. Specifically, the first synthetic property value of the first composite material is determined based on the first blending ratio of the constituent materials included in the first composite material and the property value corresponding to the constituent material included in the first composite material. Calculated. For example, the first composite characteristic value is determined based on the first compounding ratio of the first compounding data shown in FIG. 13 and the first characteristic value data (not shown) corresponding to the constituent materials included in the first composite material. It is calculated by performing an operation that combines characteristic values of value data. Note that the first characteristic value data refers to data that associates the material name of the constituent material contained in the first composite material with the characteristic value of the constituent material, and this first characteristic value data is obtained in advance. shall be taken as a thing. For example, focusing on resin D shown in FIG. 13, the first characteristic value data is data that associates the material name of resin D with the characteristic value of resin D.

Further, the value of the first categorical variable is input based on the first combination data shown in FIG. 13. Similarly, the value of the first irradiation amount is also input based on the first formulation data shown in FIG. Here, the value of the first categorical variable is the value of the categorical variable for the first composite material, and the value of the first irradiation amount is the value of the irradiation amount for the first composite material.

For example, the first composite material specified by the formulation number "ID100" has a filler volume ratio of "0.6", a maleic anhydride modification amount of "0.3", a crystal content of "27", and a vinyl acetate base content. is "32" and the filler surface area is "643". In addition, for the first composite material specified by the formulation number "ID100", the first categorical variable "CAT1" is "1", "CAT2" is "0", "CAT3" is "0", and "CAT4". is "1", and the first irradiation amount is "0".

Next, the material name and first mixture ratio included in the first formulation data shown in FIG. 13, the first composite characteristic value, the value of the first categorical variable, and the value of the first irradiation dose shown in FIG. 14 are converted into an approximation function. By inputting the information, the value of the physical quantity for the first composite material is estimated. As a result, the estimated physical quantity values are output as shown in the table of FIG. For example, in FIG. 15, it can be seen that for the first composite material specified by the formulation number "ID100", a value (estimated value) of tensile strength of "13.15" is output.

In this way, according to this specific example, it is possible to output a value of a physical quantity that is highly likely to be realized for the first composite material to be evaluated whose correspondence with the value of a physical quantity is unknown.

In particular, in this specific example, not only the categorical variables but also the composition information including the material names and composition ratios of the constituent materials, the first composite characteristic value, and the first irradiation amount (process condition value) are used as input parameters of the approximation function. are doing. Therefore, according to this specific example, the number of types of input parameters for the approximation function increases, and as a result, the accuracy of estimating the value of the physical quantity can be improved.

That is, for example, when the input parameter is "x" and the output parameter is "y", the approximation function is represented by the function "f" where y=f(x). At this time, when considering generating an approximation function by machine learning, it is often possible to obtain a highly accurate approximation function by increasing the number of types of input parameters "x". Therefore, in this specific example, the input parameter "x" is not only a categorical variable, but also a first synthetic characteristic value, composition information including "material name" and "composition ratio", and a first irradiation dose (process condition). value) is used. As a result, according to this specific example, there are more types of input parameters than when the input parameters are only categorical variables. As a result, according to this specific example, it is possible to generate a more accurate approximation function, and therefore the estimation accuracy of the value of the physical quantity can be improved.

<Verification of effectiveness>
In the following, a verification result showing that according to the second embodiment, the value of a physical quantity corresponding to a composite material whose correspondence with the value of a physical quantity is unknown can be estimated with high accuracy will be explained.

FIG. 16A is a graph showing the results of estimating the initial tensile strength of the first composite material using an approximation function that does not use categorical variables as input parameters. In FIG. 16(a), the horizontal axis shows actual measured values, while the vertical axis shows predicted values. As shown in FIG. 16(a), the value of ^R2 , which is the coefficient of determination, is "0.8357".

Against this. FIG. 16(b) is a graph showing the results of estimating the initial tensile strength of the first composite material using an approximation function using categorical variables as input parameters. In FIG. 16(b), the horizontal axis shows actual measured values, while the vertical axis shows predicted values. As shown in FIG. 16(b), the value of ^R2 , which is the coefficient of determination, is "0.8722".

At this time, considering that the larger the coefficient of determination is, the closer the predicted value is to the actual measured value, if the initial tensile strength of the first composite material is estimated using an approximation function that uses categorical variables as input parameters, It can be seen that the accuracy of estimating the initial tensile strength is improved compared to when estimating the initial tensile strength of the first composite material using an approximation function that does not use variables.

Therefore, from the verification results in FIGS. 16(a) and 16(b), according to the technical idea of the second embodiment, the value of the physical quantity corresponding to the first composite material whose correspondence with the value of the physical quantity is unknown. (For example, the value of initial tensile strength) can be estimated with high accuracy. In other words, the technique in this second embodiment of estimating the initial tensile strength of the first composite material using an approximation function using categorical variables as input parameters from the verification results in FIGS. 16(a) and 16(b) It can be seen that this idea is a useful technical idea in that the value of the physical quantity corresponding to the first composite material whose correspondence with the value of the physical quantity is unknown can be estimated with high accuracy.

The invention made by the present inventor has been specifically explained based on the embodiments thereof, but the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the gist thereof. Needless to say.

100 Physical quantity estimation device 101 CPU
102 ROM
103 RAM
104 Display 105 Keyboard 106 Mouse 107 Communication board 108 Removable disk device 109 CD/DVD-ROM device 110 Printer 111 Scanner 112 Hard disk device 113 Bus 201 Operating system 202 Program group 203 File group
301 Input section 301A Input section 301B Input section 302 Characteristic value data extraction section 302A First characteristic value data extraction section 302B Characteristic value data extraction section 303 Combined characteristic value calculation section 303A First combined characteristic value calculation section 303B Combined characteristic value calculation section 304 Synthesis-related data generation unit 304A Additional data generation unit 305 Approximation function generation unit 306 Physical quantity estimation unit 307 Output unit 308 Data storage unit 309A Communication unit 309B Communication unit 310A Data storage unit 310B Data storage unit 400 Physical quantity estimation device 500 Approximation function generation device

Claims (20)

A physical quantity estimation system for estimating the value of a physical quantity for a composite material containing two or more materials belonging to a plurality of different materials as constituent materials, the system comprising:
When inputting the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of a first composite material whose value of the physical quantity is unknown, the value of the physical quantity for the first composite material is input. an approximation function generation unit that generates an approximation function that outputs a value;
a physical quantity estimation unit that estimates the value of the physical quantity for the first composite material based on the value of the first categorical variable and the approximation function;
A physical quantity estimation system. The physical quantity estimation system according to claim 1,
The first composite material includes a flame retardant,
The first categorical variable is
a first variable indicating the presence or absence of a first surface treatment of the flame retardant;
a second variable indicating the presence or absence of a second surface treatment of the flame retardant;
a third variable indicating the presence or absence of a first material constituting the flame retardant;
a fourth variable indicating the presence or absence of a second material constituting the flame retardant;
Physical quantity estimation system, including The physical quantity estimation system according to claim 2,
The first surface treatment is a silane coupling treatment,
The second surface treatment is fatty acid treatment,
The first material is magnesium hydroxide,
The physical quantity estimation system, wherein the second material is aluminum hydroxide. The physical quantity estimation system according to claim 1,
The physical quantity estimation system calculates the first composite material based on a first blending ratio of constituent materials included in the first composite material and a first characteristic value corresponding to the constituent material included in the first composite material. comprising a composite characteristic value calculation unit that calculates a first composite characteristic value of
When the approximation function generation unit receives the first composite characteristic value and the value of the first categorical variable, it generates an approximation function that outputs the value of the physical quantity for the first composite material,
A physical quantity estimation system, wherein the physical quantity estimation unit estimates the value of the physical quantity for the first composite material based on the first composite characteristic value, the value of the first categorical variable, and the approximation function. The physical quantity estimation system according to claim 1,
When the approximation function generation unit receives the value of the first categorical variable and the first process condition value, it generates an approximation function that outputs the value of the physical quantity for the first composite material,
A physical quantity estimation system, wherein the physical quantity estimation unit estimates a value of the physical quantity for the first composite material based on a value of the first categorical variable, the first process condition value, and the approximation function. The physical quantity estimation system according to claim 5,
A physical quantity estimation system, wherein the first process condition value is a value of radiation irradiation amount in a resin crosslinking process. The physical quantity estimation system according to claim 1,
A physical quantity estimation system, wherein the physical quantity is elongation or tensile strength. An approximate function generation device that is a component of a physical quantity estimation system that estimates the value of a physical quantity for a composite material that includes two or more materials belonging to a plurality of different materials as constituent materials,
When inputting the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of a first composite material whose value of the physical quantity is unknown, the value of the physical quantity for the first composite material is input. An approximation function generation device comprising an approximation function generation unit that generates an approximation function that outputs a value. The approximate function generation device according to claim 8,
The approximation function generation unit generates an approximation function that outputs the value of the physical quantity for the first composite material when the first composite characteristic value and the value of the first categorical variable are input,
The first composite characteristic value is a value calculated based on a first blending ratio of constituent materials included in the first composite material and a first characteristic value corresponding to the constituent materials included in the first composite material. An approximate function generator. The approximate function generation device according to claim 8,
An approximation function generation device, wherein the approximation function generation unit generates an approximation function that outputs a value of the physical quantity for the first composite material when the value of the first categorical variable and the first process condition value are input. A program that causes a computer to execute a process of estimating the value of a physical quantity for a composite material containing two or more materials belonging to a plurality of different materials as constituent materials, the program comprising:
When inputting the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of the first composite material whose value of the physical quantity is unknown, the value of the physical quantity for the first composite material is input. A program that includes approximate function generation processing that generates an approximate function that outputs a value. A computer-readable recording medium recording the program according to claim 11. A physical quantity estimation device that is a component of a physical quantity estimation system that estimates the value of a physical quantity for a composite material containing two or more materials belonging to a plurality of different materials as constituent materials,
Based on the approximation function and the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of the first composite material whose value of the physical quantity is unknown, the first composite material comprising a physical quantity estimation unit that estimates the value of the physical quantity for
The approximation function is a physical quantity estimating device that is a function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable is input. The physical quantity estimating device according to claim 13,
A first synthetic characteristic value of the first composite material based on a first blending ratio of constituent materials included in the first composite material and a first characteristic value corresponding to the constituent material contained in the first composite material. It has a composite characteristic value calculation unit that calculates
The physical quantity estimation unit estimates the value of the physical quantity for the first composite material based on the first composite characteristic value, the value of the first categorical variable, and the approximation function,
The approximation function is a physical quantity estimating device that is a function that outputs the value of the physical quantity for the first composite material when the first composite characteristic value and the value of the first categorical variable are input. The physical quantity estimating device according to claim 13,
The approximation function is a function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable and the first process condition value are input,
The physical quantity estimating unit is a physical quantity estimating device, wherein the physical quantity estimation unit estimates the value of the physical quantity for the first composite material based on the value of the first categorical variable, the first process condition value, and the approximation function. A program that causes a computer to execute a process of estimating the value of a physical quantity for a composite material containing two or more materials belonging to a plurality of different materials as constituent materials, the program comprising:
Based on the approximation function and the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of the first composite material whose value of the physical quantity is unknown, the first composite material comprising a physical quantity estimation process for estimating the value of the physical quantity for,
The approximation function is a program that is a function that outputs the value of the physical quantity for the first composite material when the value of the first categorical variable is input. A computer-readable recording medium recording the program according to claim 16. A physical quantity estimation method in which a computer estimates the value of a physical quantity for a composite material containing two or more materials belonging to a plurality of different materials as constituent materials, the method comprising:
When inputting the value of a first categorical variable consisting of a digital variable associated with a factor that affects the value of the physical quantity of a first composite material whose value of the physical quantity is unknown, the value of the physical quantity for the first composite material is input. an approximation function generation step in which an approximation function generation unit of the computer generates an approximation function that outputs a value;
a physical quantity estimating step in which a physical quantity estimating unit of a computer estimates the value of the physical quantity for the first composite material based on the value of the first categorical variable and the approximation function;
A physical quantity estimation method comprising: The physical quantity estimation method according to claim 18,
In the physical quantity estimation method, a synthetic characteristic value calculation unit of a computer calculates a first blending ratio of constituent materials included in the first composite material and a first characteristic value corresponding to the constituent materials included in the first composite material. a synthetic characteristic value calculation step of calculating the first synthetic characteristic value of the first composite material based on the
In the approximation function generation step, when the approximation function generation unit of the computer inputs the first composite characteristic value and the value of the first categorical variable, the approximation function generation unit outputs the value of the physical quantity for the first composite material. generate a function,
In the physical quantity estimating step, the physical quantity estimation unit of the computer estimates the value of the physical quantity for the first composite material based on the first composite characteristic value, the value of the first categorical variable, and the approximation function. , physical quantity estimation method. The physical quantity estimation method according to claim 18,
In the approximation function generation step, when the approximation function generation unit of the computer receives the value of the first categorical variable and the first process condition value, it generates an approximation function that outputs the value of the physical quantity for the first composite material. generate,
In the physical quantity estimating step, the physical quantity estimation unit of the computer estimates the value of the physical quantity for the first composite material based on the value of the first categorical variable, the first process condition value, and the approximation function. , physical quantity estimation method.

Images