Material design apparatus, material design method, and material design program

Patent metadata

Field	Value
Publication number	US-12579331-B2
Application number	US-202017753402-A
Country	US
Kind code	B2
Filing date	Sep 1, 2020
Priority date	Sep 6, 2019
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

A material design apparatus includes a learned model that has learned a correspondence between input information about a blend proportion of a monomer and output information about physical property values of a polymer by machine learning. Each unit of the material design apparatus is configured to: receive as input a blend proportion range of at least one monomer; receive required ranges of physical property values of a polymer; generate a comprehensive analysis point of a polymer polymerized from multiple monomers, the multiple monomers including, within the blend proportion range, at least one monomer of which the blend proportion range is input; input the generated comprehensive analysis point into the learned model to calculate physical property values of a polymer, to create a data set, and to store the created data set; and select a polymer within the required ranges of the physical property values from the data set.

First claim

The invention claimed is:

1 . A material design apparatus for designing a polymer polymerized from multiple types of monomers, the material design apparatus comprising:
a learned model that has learned a correspondence between input information about a blend proportion of a monomer and output information about a plurality of physical property values of a polymer by machine learning, a blend proportion range input unit configured to receive as input a blend proportion range of at least one monomer, a required physical property input unit configured to receive as input required ranges of a plurality of physical property values of a polymer, a comprehensive analysis point generation unit configured to generate a comprehensive analysis point of a polymer polymerized from multiple monomers, the multiple monomers including, within the blend proportion range, at least one monomer of which the blend proportion range is input, a comprehensive analysis point-polymer physical property value storage unit configured to input the generated comprehensive analysis point into the learned model to calculate a plurality of physical property values of a polymer, to create a data set in which the comprehensive analysis point and the calculated physical property values of the polymer are linked, and to store the created data set, and a filter unit configured to select a polymer within the required ranges of the physical property values input in the required physical property input unit from the data set.

2 . The material design apparatus according to claim 1 , wherein in the blend proportion range input unit, a number of monomers used for polymerization is input, and the number of monomers used for polymerization is limited, and wherein a comprehensive analysis point of a polymer polymerized using a limited number of monomers is generated.

3 . The material design apparatus according to claim 1 , wherein in the blend proportion range input unit, at least one monomer is input as essential for polymerization from among monomers of which the blend proportion range is input, and wherein a comprehensive analysis point of a polymer polymerized from multiple monomers including an essential monomer within the blend proportion range is generated.

4 . A material design apparatus for designing a graft polymer polymerized in two stages from multiple types of monomers, the material design apparatus comprising:
a learned model that has learned a correspondence between input information about a blend proportion of a monomer and output information about a plurality of physical property values of a polymer by machine learning, a blend proportion range input unit configured to receive as input a blend proportion range of at least one monomer, a required physical property input unit configured to receive as input required ranges of a plurality of physical property values of a polymer, a first stage comprehensive analysis point generation unit configured to select at least one first stage monomer used for first stage polymerization from among monomers of which the blend proportion range is input, and to generate a comprehensive analysis point of a main chain polymer polymerized using multiple monomers including the at least one first stage monomer within the blend proportion range, a second stage monomer proposal unit configured to propose at least one second stage monomer used for a second stage polymerization with the main chain polymer based on a first stage comprehensive analysis point, an integrated comprehensive analysis point generation unit configured to generate an integrated comprehensive analysis point of a graft polymer obtained by polymerizing the second stage monomer with the main chain polymer, a comprehensive analysis point-polymer physical property value storage unit configured to input the integrated comprehensive analysis point into the learned model to calculate a plurality of physical property values of a graft polymer, to create a data set in which the integrated comprehensive analysis point and the calculated physical property values of the graft polymer are linked, and to store the created data set, and a filter unit configured to select a graft polymer within the required ranges of the physical property values input in the required physical property input unit from the data set.

5 . The material design apparatus according to claim 4 , wherein in the blend proportion range input unit, a number of monomers used for polymerization is input, and the number of monomers used for polymerization is limited, and wherein an integrated comprehensive analysis point of a graft polymer polymerized using a limited number of monomers is generated.

6 . The material design apparatus according to claim 4 , wherein in the blend proportion range input unit, at least one monomer is input as essential for polymerization from among monomers of which the blend proportion range is input, and wherein in the integrated comprehensive analysis point generation unit, a comprehensive analysis point of a graft polymer polymerized from multiple monomers including an essential monomer within the blend proportion range is generated.

7 . The material design apparatus according to claim 4 , wherein in the blend proportion range input unit, at least one monomer is input as essential for first stage polymerization from among monomers of which the blend proportion range is input, and wherein in the first stage comprehensive analysis point generation unit, a comprehensive analysis point of a main chain polymer polymerized from multiple monomers including an essential monomer within the blend proportion range is generated.

8 . A material design method for designing a polymer polymerized from multiple types of monomers, the material design method comprising:
creating a learned model that has learned a correspondence between input information about a blend proportion of a monomer and output information about a plurality of physical property values of a polymer by machine learning, receiving as input a blend proportion range of at least one monomer, receiving as input required ranges of a plurality of physical property values of a polymer, generating a comprehensive analysis point of a polymer polymerized using, within the blend proportion range, at least one monomer of which the blend proportion range is input, inputting the comprehensive analysis point into the learned model to calculate a plurality of physical property values of a polymer, creating a data set in which the comprehensive analysis point and the calculated physical property values of the polymer are linked, and storing the created data set in a comprehensive analysis point-polymer physical property value storage unit, and selecting a polymer within the required ranges of the physical property values from the data set.

9 . A non-transitory computer-readable recording medium storing a material design program for designing a polymer polymerized from multiple types of monomers, the material design program causing a computer to implement functions comprising:
creating a learned model that has learned a correspondence between input information about a blend proportion of a monomer and output information about a plurality of physical property values of a polymer by machine learning, receiving as input a blend proportion range of at least one monomer, receiving as input required ranges of a plurality of physical property values of a polymer, generating a comprehensive analysis point of a polymer polymerized using, within the blend proportion range, at least one monomer of which the blend proportion range is input, inputting the comprehensive analysis point into the learned model to calculate a plurality of ph

Assignees

• Showa Denko Kk
• Resonac Corp

Inventors

• Aonuma Naoto
• Takemoto Shimpei
• Takeda Eriko
• Okuno Yoshishige
• Kuramoto Hiroki
• Kawaguchi Yasuaki

Classifications

• G06F30/27

  using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model · CPC title

• G16C20/30

  Prediction of properties of chemical compounds, compositions or mixtures · CPC title

• G16C20/70

  Machine learning, data mining or chemometrics · CPC title

• G06F30/10 Primary

  Geometric CAD · CPC title

• G16C60/00 Primary

  Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation · CPC title