서브메뉴
검색
Rigorously Quantifying Uncertainties for Transport Phenomena in Molecular Simulations- [electronic resource]
Rigorously Quantifying Uncertainties for Transport Phenomena in Molecular Simulations- [electronic resource]
상세정보
- 자료유형
- 학위논문파일 국외
- 최종처리일시
- 20240214101916
- ISBN
- 9798380478304
- DDC
- 620
- 저자명
- Li, Yuanhao.
- 서명/저자
- Rigorously Quantifying Uncertainties for Transport Phenomena in Molecular Simulations - [electronic resource]
- 발행사항
- [S.l.]: : Carnegie Mellon University., 2023
- 발행사항
- Ann Arbor : : ProQuest Dissertations & Theses,, 2023
- 형태사항
- 1 online resource(122 p.)
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-04, Section: B.
- 주기사항
- Advisor: Wang, Gerald J.
- 학위논문주기
- Thesis (Ph.D.)--Carnegie Mellon University, 2023.
- 사용제한주기
- This item must not be sold to any third party vendors.
- 초록/해제
- 요약The field of computational materials science faces various challenges in data processing, including dealing with high-dimensional parameter spaces and error analysis. Uncertainty quantification has become crucial for interpreting the results of materials simulations. We begin by discussing the fundamentals and challenges associated with addressing uncertainties in molecular-dynamics (MD) simulations. Subsequently, we address two problems at the heart of uncertainty analysis for MD simulations. In the first problem, under the assumption that we have a large dataset consisting of numerous statistically independent MD datasets (each of which can be used to estimate a quantity of engineering interest via, e.g., regression analysis), we investigate the statistical confidence we can build using the large dataset as a whole. In the second problem, we study physical settings in which the assumption underlying the first problem is likely to fail, namely, settings in which nominally independent MD datasets are not in fact independent. Both problems have significant relevance for the interpretation of simulations of many nanoscale phenomena (e.g., confined fluid diffusion or heat transfer at fluid-solid interfaces).We discuss an approach for conducting regression analysis on time series data designed to circumvent the challenges posed by the first problem. Additionally, we explore an approach rooted in thermodynamic principles that accelerates the decorrelation between consecutive MD measurements, offering a solution to the second problem. Leveraging these approaches, we study thermal transport properties at a fluid-solid interface. Finally, we propose a Heteroscedastic Gaussian Process Regression workflow to model fluid self-diffusion coefficient as a function of thermodynamic conditions.
- 일반주제명
- Engineering.
- 일반주제명
- Civil engineering.
- 기타저자
- Carnegie Mellon University Civil and Environmental Engineering
- 기본자료저록
- Dissertations Abstracts International. 85-04B.
- 기본자료저록
- Dissertation Abstract International
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008240612s2023 us |||||||||||||||c||eng d■001000016935304
■00520240214101916
■006m o d
■007cr#unu||||||||
■020 ▼a9798380478304
■035 ▼a(MiAaPQ)AAI30688305
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a620
■1001 ▼aLi, Yuanhao.▼0(orcid)0000-0003-0688-0910
■24510▼aRigorously Quantifying Uncertainties for Transport Phenomena in Molecular Simulations▼h[electronic resource]
■260 ▼a[S.l.]:▼bCarnegie Mellon University. ▼c2023
■260 1▼aAnn Arbor :▼bProQuest Dissertations & Theses, ▼c2023
■300 ▼a1 online resource(122 p.)
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-04, Section: B.
■500 ▼aAdvisor: Wang, Gerald J.
■5021 ▼aThesis (Ph.D.)--Carnegie Mellon University, 2023.
■506 ▼aThis item must not be sold to any third party vendors.
■520 ▼aThe field of computational materials science faces various challenges in data processing, including dealing with high-dimensional parameter spaces and error analysis. Uncertainty quantification has become crucial for interpreting the results of materials simulations. We begin by discussing the fundamentals and challenges associated with addressing uncertainties in molecular-dynamics (MD) simulations. Subsequently, we address two problems at the heart of uncertainty analysis for MD simulations. In the first problem, under the assumption that we have a large dataset consisting of numerous statistically independent MD datasets (each of which can be used to estimate a quantity of engineering interest via, e.g., regression analysis), we investigate the statistical confidence we can build using the large dataset as a whole. In the second problem, we study physical settings in which the assumption underlying the first problem is likely to fail, namely, settings in which nominally independent MD datasets are not in fact independent. Both problems have significant relevance for the interpretation of simulations of many nanoscale phenomena (e.g., confined fluid diffusion or heat transfer at fluid-solid interfaces).We discuss an approach for conducting regression analysis on time series data designed to circumvent the challenges posed by the first problem. Additionally, we explore an approach rooted in thermodynamic principles that accelerates the decorrelation between consecutive MD measurements, offering a solution to the second problem. Leveraging these approaches, we study thermal transport properties at a fluid-solid interface. Finally, we propose a Heteroscedastic Gaussian Process Regression workflow to model fluid self-diffusion coefficient as a function of thermodynamic conditions.
■590 ▼aSchool code: 0041.
■650 4▼aEngineering.
■650 4▼aEnvironmental engineering.
■650 4▼aCivil engineering.
■653 ▼aMolecular-dynamics
■653 ▼aMaterials simulations
■653 ▼aThermodynamic conditions
■653 ▼aFluid-solid interface
■690 ▼a0537
■690 ▼a0543
■690 ▼a0775
■71020▼aCarnegie Mellon University▼bCivil and Environmental Engineering.
■7730 ▼tDissertations Abstracts International▼g85-04B.
■773 ▼tDissertation Abstract International
■790 ▼a0041
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T16935304▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
■980 ▼a202402▼f2024
