본문

서브메뉴

Efficient Deep Learning Models for Physics Simulation
Efficient Deep Learning Models for Physics Simulation
Efficient Deep Learning Models for Physics Simulation

Detailed Information

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  
Factorized attention
키워드  
Deep learning
키워드  
Diffusion probabilistic models
키워드  
Graph neural networks
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.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Подробнее информация.

    • Бронирование
    • не существует
    • моя папка
    • Первый запрос зрения
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    материал
    Reg No. Количество платежных Местоположение статус Ленд информации
    TF19023 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Бронирование доступны в заимствований книги. Чтобы сделать предварительный заказ, пожалуйста, нажмите кнопку бронирование

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.