http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
기계학습을 이용한 신규 Double Perovskite 스크리닝
김준철(Joonchul Kim),민경민(Kyoungmin Min) 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.12
Double perovskite structures have brought a lot of attention due to their great potentials for applications of batteries, lighting-devices, and energy harvesting materials. In this study, machine learning algorithm is employed to search for new stable double perovskite materials. First, the materials properties are adopted from well-established Materials Project database to develop a prediction model for the formation energy and the convex hull energy. Then, the bagging and boosting based algorithms are implemented to train the database for regression as well as classification model and their prediction accuracy is compared. For prediction of the formation energy, it’s R2 and RMSE value reaches to 0.97 and 0.2020 eV/atom. In addition, the classification accuracy for the convex hull energy shows 0.76 with F1-score of 0.763. Finally, trained machine learning model is applied to the whole chemical space of the double perovskite structures and it exhibits that 8,613 structures are potentially stable to be synthesized. In the meanwhile, 25,062 and 19,766 structures are shown to be metastable and instable, respectively.
Molecular Dynamics simulation을 이용한 3D printed zeolite 구조체의 미시적, 거시적 기계적 거동 분석
김준철(Joonchul Kim),민경민(Kyoungmin Min) 대한기계학회 2023 대한기계학회 춘추학술대회 Vol.2023 No.11
Mechanical properties are important properties from atomic unit structures to micro-level structures. Discovering the correlation between structural information and mechanical reactions can accelerate the structure design with desired mechanical properties. By establishing a link between mechanical properties and structural information, we prove the possibility of purposeful structural design. We focus on uncovering resemblances in mechanical behavior between atomistic arrangements and 3D-printed zeolite structures. Employing molecular dynamics simulations, we validate the emergence of similar mechanical responses at both atomic and macroscopic scales. 3D printed structure using thermoplastic polyurethane (TPU) filaments can reflect the response of microstructural-level simulations, which can link between experiment and theory. These results demonstrate that the design can realize meta-materials with ideal mechanical reactions based on theoretical results of atomic structures.
데이터 최적화 기법을 도입한 기계적 물성치 예측 모델과 새로운 2차원 소재의 발견
이인효(Inhyo Lee),김준철(Joonchul Kim),김태현(Taehyun Kim),민경민(Kyoungmin Min) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
Two-dimensional (2D) materials are attractive materials. Many studies are being conducted because of their unique characteristics. However, there is lack of information about properties of 2D materials. Therefore, this study attempted to solve this problem by developing a machine learning (ML) model that predicts mechanical properties of 2D materials. In addition, a 2D materials generation framework was developed using a classification model and a deep learning-based generative model. ML model to predict mechanical properties is trained from existing 2D database and reduces the uncertainty of prediction through data optimization techniques. Potential 2D materials are discovered through screening processes such as measuring structure and atomic similarities. We believe that the developing of ML model and framework for finding new 2D materials could open a new chapter in material science
데이터 최적화 기법을 도입한 기계적 물성치 예측 모델과 새로운 2차원 소재의 발견
이인효(Inhyo Lee),김준철(Joonchul Kim),김태현(Taehyun Kim),민경민(Kyoungmin Min) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
Two-dimensional (2D) materials are attractive materials. Many studies are being conducted because of their unique characteristics. However, there is lack of information about properties of 2D materials. Therefore, this study attempted to solve this problem by developing a machine learning (ML) model that predicts mechanical properties of 2D materials. In addition, a 2D materials generation framework was developed using a classification model and a deep learning-based generative model. ML model to predict mechanical properties is trained from existing 2D database and reduces the uncertainty of prediction through data optimization techniques. Potential 2D materials are discovered through screening processes such as measuring structure and atomic similarities. We believe that the developing of ML model and framework for finding new 2D materials could open a new chapter in material science