| 학회 |
한국재료학회 |
| 학술대회 |
2019년 가을 (10/30 ~ 11/01, 삼척 쏠비치 호텔&리조트) |
| 권호 |
25권 2호 |
| 발표분야 |
특별심포지엄2. 재료공학에 적용 가능한 인공지능 기술 심포지엄-오거나이저:이승철(포항공대) |
| 제목 |
A machine learning approach to establish microstructure-property linkages |
| 초록 |
Microstructure based simulations are one of the most accurate ways in predicting and understanding how microstructural variables relate to properties of a material. Unfortunately, the simulations are only qualitatively used in optimizing materials microstructures or in establishing comprehensive structure-property linkages due to computational cost restrains. This research demonstrates that by utilizing machine learning techniques, one can alleviate the computational cost restrain and construct structure-property linkages for a wide range of microstructures using only a small number of full-field simulation results. Furthermore, with the implementation of Bayesian optimization, one can possibly identify microstructures that exhibit most desirable properties using even smaller number of full-field simulations. |
| 저자 |
정재면1, 윤재익1, 박형근1, 김진유2, 김형섭1
|
| 소속 |
1포항공과대, 2POSCO |
| 키워드 |
Microstructure; Machine learning; Gaussian process regression; Optimization
|
| E-Mail |
|
