서브메뉴
검색
Computational Materials Design: Integrating Physics With Optimization and Machine Learning- [electronic resource]
Computational Materials Design: Integrating Physics With Optimization and Machine Learning- [electronic resource]
상세정보
- 자료유형
- 학위논문파일 국외
- 최종처리일시
- 20240214101911
- ISBN
- 9798380381543
- DDC
- 660
- 저자명
- Yin, Xiangyu.
- 서명/저자
- Computational Materials Design: Integrating Physics With Optimization and Machine Learning - [electronic resource]
- 발행사항
- [S.l.]: : Carnegie Mellon University., 2023
- 발행사항
- Ann Arbor : : ProQuest Dissertations & Theses,, 2023
- 형태사항
- 1 online resource(135 p.)
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
- 주기사항
- Advisor: Gounaris, Chrysanthos.
- 학위논문주기
- Thesis (Ph.D.)--Carnegie Mellon University, 2023.
- 사용제한주기
- This item must not be sold to any third party vendors.
- 초록/해제
- 요약Computational materials discovery involves a complex interplay between physics-based modeling, data-driven machine learning, and optimization. Despite fast-growing advancements, there exist challenges in efficiency, accuracy, scalability, generalizability, and interpretability. Many of those challenges arise from the lack of exploiting problem structures and constraints or handling the data bias in the data. In general, computational materials discovery involves generation, representation, evaluation, and search. Search is the bottleneck of many challenges mentioned before, but systematic search strategies have yet to be investigated thoroughly. On the other hand, there are also some engineering challenges in building the computational workflow regarding standardization and automation. Moreover, being a highly interdisciplinary and inherently multiscale field of study, the success of computational materials discovery requires joint efforts from materials scientists, chemical engineers, process engineers, etc., resulting in different scientific languages and knowledge barriers that hinder efficient collaboration. In this thesis, we aimed to integrate physics into computational methodologies to improve problem-solving and advance materials discovery. We focus on the search aspect of computational materials discovery and propose novel optimization methodologies to tackle many scientific challenges. Besides, this thesis also involves developing scientific software to tackle some of the engineering challenges mentioned above by developing good computational tools and workflows that align with domain knowledge and are robust, automated, and easy to use.
- 일반주제명
- Chemical engineering.
- 일반주제명
- Applied physics.
- 일반주제명
- Materials science.
- 키워드
- Machine learning
- 기타저자
- Carnegie Mellon University Chemical Engineering
- 기본자료저록
- Dissertations Abstracts International. 85-03B.
- 기본자료저록
- Dissertation Abstract International
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008240612s2023 us |||||||||||||||c||eng d■001000016935260
■00520240214101911
■006m o d
■007cr#unu||||||||
■020 ▼a9798380381543
■035 ▼a(MiAaPQ)AAI30686851
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a660
■1001 ▼aYin, Xiangyu.▼0(orcid)0000-0003-2868-1728
■24510▼aComputational Materials Design: Integrating Physics With Optimization and Machine Learning▼h[electronic resource]
■260 ▼a[S.l.]:▼bCarnegie Mellon University. ▼c2023
■260 1▼aAnn Arbor :▼bProQuest Dissertations & Theses, ▼c2023
■300 ▼a1 online resource(135 p.)
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-03, Section: B.
■500 ▼aAdvisor: Gounaris, Chrysanthos.
■5021 ▼aThesis (Ph.D.)--Carnegie Mellon University, 2023.
■506 ▼aThis item must not be sold to any third party vendors.
■520 ▼aComputational materials discovery involves a complex interplay between physics-based modeling, data-driven machine learning, and optimization. Despite fast-growing advancements, there exist challenges in efficiency, accuracy, scalability, generalizability, and interpretability. Many of those challenges arise from the lack of exploiting problem structures and constraints or handling the data bias in the data. In general, computational materials discovery involves generation, representation, evaluation, and search. Search is the bottleneck of many challenges mentioned before, but systematic search strategies have yet to be investigated thoroughly. On the other hand, there are also some engineering challenges in building the computational workflow regarding standardization and automation. Moreover, being a highly interdisciplinary and inherently multiscale field of study, the success of computational materials discovery requires joint efforts from materials scientists, chemical engineers, process engineers, etc., resulting in different scientific languages and knowledge barriers that hinder efficient collaboration. In this thesis, we aimed to integrate physics into computational methodologies to improve problem-solving and advance materials discovery. We focus on the search aspect of computational materials discovery and propose novel optimization methodologies to tackle many scientific challenges. Besides, this thesis also involves developing scientific software to tackle some of the engineering challenges mentioned above by developing good computational tools and workflows that align with domain knowledge and are robust, automated, and easy to use.
■590 ▼aSchool code: 0041.
■650 4▼aChemical engineering.
■650 4▼aApplied physics.
■650 4▼aMaterials science.
■653 ▼aComputational materials
■653 ▼aPhysics-based modeling
■653 ▼aMachine learning
■653 ▼aEngineering challenges
■653 ▼aScientific software
■690 ▼a0542
■690 ▼a0794
■690 ▼a0800
■690 ▼a0215
■71020▼aCarnegie Mellon University▼bChemical Engineering.
■7730 ▼tDissertations Abstracts International▼g85-03B.
■773 ▼tDissertation Abstract International
■790 ▼a0041
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T16935260▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
■980 ▼a202402▼f2024
