US 12,190,032 B2
Material design device, material design method, and material design program
Katsuki Okuno, Tokyo (JP); Takuya Minami, Tokyo (JP); Shimpei Takemoto, Tokyo (JP); Eriko Takeda, Tokyo (JP); Yoshishige Okuno, Tokyo (JP); and Masamichi Kitano, Tokyo (JP)
Assigned to Showa Denko K.K., Tokyo (JP)
Appl. No. 17/289,277
Filed by SHOWA DENKO K.K., Tokyo (JP)
PCT Filed Oct. 30, 2019, PCT No. PCT/JP2019/042466
§ 371(c)(1), (2) Date Apr. 28, 2021,
PCT Pub. No. WO2020/090848, PCT Pub. Date May 7, 2020.
Claims priority of application No. 2018-204439 (JP), filed on Oct. 30, 2018.
Prior Publication US 2021/0397769 A1, Dec. 23, 2021
Int. Cl. G06F 30/27 (2020.01); C22C 1/00 (2023.01); G06F 113/26 (2020.01)
CPC G06F 30/27 (2020.01) [C22C 1/00 (2013.01); G06F 2113/26 (2020.01)] 30 Claims
OG exemplary drawing
 
1. A material design device for designing a material to be designed including a material composed of a plurality of compositions or a material produced by combining a plurality of production conditions,
the material design device comprising:
an information display unit to display an input screen to prompt a user to input a maximum value and a minimum value for a design condition, wherein the maximum value and the minimum value define the specified range of the design condition, wherein the design condition includes a material composition and a production condition;
a learned model which is trained with a correspondence between user input information including the design condition and output information including a material property value;
a comprehensive prediction point generation unit configured to generate a plurality of comprehensive prediction points within the specified range of the design condition defined by the maximum value and the minimum value inputted using the input screen, wherein the plurality of comprehensive prediction points are input into the learned model to associate the material property value with at least one of the comprehensive prediction points;
a design condition-material property table configured to store data sets in which the material property value calculated by inputting the comprehensive prediction points generated by the comprehensive prediction point generation unit to the learned model is associated with each point of the comprehensive prediction points;
a required property setting unit configured to set a specified range of a required property of the material to be designed;
a design condition extraction unit configured to extract data sets satisfying the required property set by the required property setting unit from the design condition-material property table;
the information display unit configured to display the requested property and the range of the composition related to the data sets extracted by the design condition extraction unit;
a design condition adjustment unit configured to adjust the range of the composition of the data set extracted by the design condition extraction unit in response to a user operation that changes the range of the composition displayed on the information display unit; and
a narrow down unit configured to narrow down from the extracted data sets satisfying the requested property to the data sets satisfying the composition adjusted by the design condition adjustment unit,
wherein the output on the information display unit is updated based on the narrowed down data.