서브메뉴
검색
Efficient Deep Learning Models for Physics Simulation
Efficient Deep Learning Models for Physics Simulation
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103057
- ISBN
- 9798288853104
- DDC
- 621
- 저자명
- Li, Zijie.
- 서명/저자
- Efficient Deep Learning Models for Physics Simulation
- 발행사항
- [Sl] : Carnegie Mellon University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 142 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-01, Section: B.
- 주기사항
- Advisor: Barati Farimani, Amir.
- 학위논문주기
- Thesis (Ph.D.)--Carnegie Mellon University, 2025.
- 초록/해제
- 요약Many natural and engineered systems are governed by partial differential equations (PDEs), spanning atomic interactions in molecular systems to large-scale cosmological dynamics. Solving these PDEs is essential for deepening our understanding of complex physical phenomena, enabling precise predictions, and guiding informed decision-making across diverse scientific and engineering fields. Numerical solvers are widely employed for simulating and predicting PDEs, particularly as many PDEs are challenging and infeasible to be solved analytically. These solvers typically discretize the continuous domain into a grid, converting differential equations into algebraic equations via methods such as finite difference, finite element, finite volume, or spectral approaches.Recent advances in machine learning, combined with the success of deep learning across many fields, have unlocked new possibilities for modeling complex sub-scale physical processes and developing efficient neural-network-based PDE solvers. In these approaches, neural networks parameterize the solution function of the target equation or approximate the solution operator itself, providing a flexible alternative to traditional solvers. Compared to numerical solvers, neural PDE solvers tend to be more tolerant of coarser discretizations and can eliminate the need for fine meshing, making them adaptable to various domains with reduced computational overhead. Additionally, since these models can exploit the patterns directly from data, knowledge of the precise underlying equations is not strictly necessary, offering a streamlined and efficient approach to complex physics simulations. In this thesis, we discuss the efforts in developing and build neural-network-based models for accurate and efficient prediction of a variety of physical systems. We first introduce Fluid Graph Networks (FGN) and Graph neural networks-Accelerated Molecular Dynamics (GAMD), two data-driven models parameterized with message passing neural networks, for efficient particle-based system simulation. We further present a series of Transformer-based models for modeling various physics phenomena including turbulent flow and global weather dynamics. The first is Operator Transformer (OFormer), which features a Transformer encoder-decoder framework that can be flexibly applied to different discretizations. To improve the scalability of the Transformer on higher dimensional problems, we next propose a axial factorized attention that greatly reduces the computational cost associated with high-dimensional grid. We further extend the proposed factorized attention mechanism to the sphere for accurate and efficient global weather forecasting. Lastly, we present a generative neural PDE solver drawing inspiration from recent progress of diffusion probabilistic models to improve the simulation robustness on turbulent time-dependent systems.
- 일반주제명
- Mechanical engineering
- 일반주제명
- Computer science
- 일반주제명
- Computational physics
- 일반주제명
- Engineering
- 키워드
- Deep learning
- 키워드
- PDE simulations
- 기타저자
- Carnegie Mellon University Mechanical Engineering
- 기본자료저록
- Dissertations Abstracts International. 87-01B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017356892
■00520260202103057
■006m o d
■007cr#unu||||||||
■020 ▼a9798288853104
■035 ▼a(MiAaPQ)AAI31932962
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a621
■1001 ▼aLi, Zijie.▼0(orcid)0000-0002-8566-7538
■24510▼aEfficient Deep Learning Models for Physics Simulation
■260 ▼a[Sl]▼bCarnegie Mellon University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a142 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-01, Section: B.
■500 ▼aAdvisor: Barati Farimani, Amir.
■5021 ▼aThesis (Ph.D.)--Carnegie Mellon University, 2025.
■520 ▼aMany natural and engineered systems are governed by partial differential equations (PDEs), spanning atomic interactions in molecular systems to large-scale cosmological dynamics. Solving these PDEs is essential for deepening our understanding of complex physical phenomena, enabling precise predictions, and guiding informed decision-making across diverse scientific and engineering fields. Numerical solvers are widely employed for simulating and predicting PDEs, particularly as many PDEs are challenging and infeasible to be solved analytically. These solvers typically discretize the continuous domain into a grid, converting differential equations into algebraic equations via methods such as finite difference, finite element, finite volume, or spectral approaches.Recent advances in machine learning, combined with the success of deep learning across many fields, have unlocked new possibilities for modeling complex sub-scale physical processes and developing efficient neural-network-based PDE solvers. In these approaches, neural networks parameterize the solution function of the target equation or approximate the solution operator itself, providing a flexible alternative to traditional solvers. Compared to numerical solvers, neural PDE solvers tend to be more tolerant of coarser discretizations and can eliminate the need for fine meshing, making them adaptable to various domains with reduced computational overhead. Additionally, since these models can exploit the patterns directly from data, knowledge of the precise underlying equations is not strictly necessary, offering a streamlined and efficient approach to complex physics simulations. In this thesis, we discuss the efforts in developing and build neural-network-based models for accurate and efficient prediction of a variety of physical systems. We first introduce Fluid Graph Networks (FGN) and Graph neural networks-Accelerated Molecular Dynamics (GAMD), two data-driven models parameterized with message passing neural networks, for efficient particle-based system simulation. We further present a series of Transformer-based models for modeling various physics phenomena including turbulent flow and global weather dynamics. The first is Operator Transformer (OFormer), which features a Transformer encoder-decoder framework that can be flexibly applied to different discretizations. To improve the scalability of the Transformer on higher dimensional problems, we next propose a axial factorized attention that greatly reduces the computational cost associated with high-dimensional grid. We further extend the proposed factorized attention mechanism to the sphere for accurate and efficient global weather forecasting. Lastly, we present a generative neural PDE solver drawing inspiration from recent progress of diffusion probabilistic models to improve the simulation robustness on turbulent time-dependent systems.
■590 ▼aSchool code: 0041.
■650 4▼aMechanical engineering
■650 4▼aComputer science
■650 4▼aComputational physics
■650 4▼aEngineering
■653 ▼aFactorized attention
■653 ▼aDeep learning
■653 ▼aDiffusion probabilistic models
■653 ▼aGraph neural networks
■653 ▼aPDE simulations
■690 ▼a0548
■690 ▼a0984
■690 ▼a0216
■690 ▼a0537
■71020▼aCarnegie Mellon University▼bMechanical Engineering.
■7730 ▼tDissertations Abstracts International▼g87-01B.
■790 ▼a0041
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17356892▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
