课程大纲
COURSE SYLLABUS
1.
课程代码/名称
Course Code/Title
MAT7087 计算流体力学与深度学习
MAT7087 Computational Fluid Dynamics and Deep Learning
2.
课程性质
Compulsory/Elective
选修 Elective
3.
课程学分/学时
Course Credit/Hours
3/48
4.
授课语
Teaching Language
中文
Chinese
5.
授课教
Instructor(s)
吴开亮副教授
Kailiang Wu, Associate Professor
6.
是否面向本科生开放
Open to undergraduates
or not
Yes
7.
先修要
Pre-requisites
If the course is open to
undergraduates, please indicate the difference.)
本科生需先修以下课程(Pre-requisites for undergraduates):
MA305 数值分析
MA305 Numerical Analysis
8.
教学目
Course Objectives
If the course is open to undergraduates, please indicate the
difference.)
掌握流体力学方程的基本数值方法、深度学习的基本知识、数据驱动的深度学习方法及其在计算
流体力学中的应用。
Master some fundamental numerical methods for fluid dynamic equations, some basic knowledge of
machine learning, data-driven deep learning methods and their applications in computational fluid
dynamics.
9.
教学方
Teaching Methods
If the course is open to undergraduates, please indicate the
difference.)
专题性质授课,从基本知识和经典方法开始介绍,并辅以最新方法和前沿课题应用
Teaching in topics, from the basic knowledge and classical methods to modern methods and frontier problems
10.
教学内
Course Contents
(如面向本科生开放,请注明区分内容。 If the course is open to undergraduates, please indicate the
difference.)
Section 1
计算流体力学简介(2 学时)
Introduction to Computational Fluid Dynamics (2-hour lectures)
Section 2
双曲型守恒律(2 学时)
Hyperbolic Conservation Laws (2-hour lectures)
Section 3
有限差分法3 学时)
Finite Difference Methods (3-hour lectures)
Section 4
黎曼问题和有限体积法(5 学时)
Riemann Problems and Finite Volume Methods (5-hour lectures)
Section 5
高分辨率格式与限制器(3 学时)
High-Resolution Schemes and Limiters (3-hour lectures)
Section 6
ENO WENO 格式(3 学时)
ENO and WENO Schemes (3-hour lectures)
Section 7
间断 Galerkin 方法(4 学时)
Discontinuous Galerkin Methods (4-hour lectures)
Section 8
高阶时间离散(3 学时)
High-Order Time Discretization (3-hour lectures)
Section 9
机器学习简介(2 学时)
Introduction to Machine Learning (2-hour lectures)
Section 10
深度神经网络(4 学时)
Deep Neural Network (4-hour lectures)
Section 11
深度学习与数据驱动方法4 学时)
Machine Learning and Data-Driven Methods (4-hour lectures)
Section 12
计算流体力学中的深度学习方法(4 学时)
Deep Learning Methods in Computational Fluid Dynamics (4-hour lectures)
Section 13
物理信息神经网络与数据驱动建模(3 学时)
Physics-Informed Neural Network and Data-Driven Modeling (3-hour
lectures)
Section 14
多维流体力学系统及其方法(6 学时)
Multidimensional Fluid Dynamic Systems and Methods (6-hour lectures)
11.
课程考
Course Assessment
1
Form of examination;
2
. grading policy
3
If the course is open to undergraduates, please indicate the difference.)
作业(50%+期末考试(50%
Assignment (50%) + Final Exam (50%)
12.
教材及其它参考资料
Textbook and Supplementary Readings
1. Eleuterio F. Toro, Riemann Solvers and Numerical Methods for Fluid Dynamics: A Practical
Introduction. Springer Science & Business Media, 2013.
2. Jan S. Hesthaven, Numerical Methods for Conservation Laws: From Analysis to Algorithms, Society
for Industrial and Applied Mathematics, 2017.
3. Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning, The MIT Press, Cambridge,
MA, USA, 2016.
4. Ke-Lin Du and Madisetti NS Swamy, Neural Networks and Statistical Learning, Springer Science &
Business Media, 2013.
5. A. Chorin and J. Marsden, A Mathematical Introduction to Fluid Mechanics, Springer-Verlag, 2000.
6. R. J. LeVeque, Finite Volume Methods for Hyperbolic Problems, Cambridge University Press, 2002.