CSE 291 (SP 2026) Physics Simulation

Welcome to CSE 291 Topics in CSE: Physics Simulation.


Homework

Syllabus

Week Tuesday Thursday
1 3/31: Introduction, ODEs 4/2: No class (instructor unavailable)
2 4/7: Dimensional Analysis 4/9: Differentials and Gradients
3 4/14: Calculus of variations 4/16: Least action principle
4 4/21: KKT conditions, Rigid motion dynamics 4/23: Rigid motion dynamics,
5 4/28:Variational integrator, Incremental potential 4/30: Incremental potential
6 5/5: Continuum Mechanics, Continuity equation 5/7: Fluid dynamics
7 5/12: Principled derivation; fluid numerics 5/14: Fluid numerics; vorticity
8 5/19: Tensor for Elasticity 5/21:
9 5/26: Tensor 5/28:Elasticity derivation
10 6/2:Elasticity, Airy function theory 6/4:Water wave
Final 6/10: No class 6/12: No class
  • Final project due
Summer break
To-be-determined lectures may be on
  • Water waves: dispersion relations
  • Singular perturbations: viscous boundary layers
  • Homogenization: multiscale modeling
  • Lie algebra: non-holonomic constraints
  • Euler--Arnold equations: coadjoint systems and vortex mechanics.

Overview

The course covers the mathematical and computational basis for various physics simulation tasks including solid mechanics and fluid dynamics. A main focus is constitutive modeling, that is, the dynamics are derived from a few universal principles of classical mechanics, such as dimensional analysis, Hamiltonian principle, maximal dissipation principle, Noether’s theorem, etc. These principles are the foundation to computational methods that can produce structure-preserving and realistic simulations.

You are expected to have basic programming and visualization skill.

Some highlight:

  • Dimensional analysis: Utilize the scaling symmetry of a physical system to estimate the expected result, to identify equivalent models, or to reduce the number of parameters.
  • Variational principles: Instead of accounting for all internal forces in a physical system, derive them by taking derivative of energy functions.
  • Incremental potential: Turn each time step into an optimization problem, encompassing dissipative systems and impulsive systems.
  • Constrained systems: Holonomic and nonholonomic constraints, quasivelocities, and the KKT conditions for inequality constraints.
  • Continuum mechanics: An exposition of tensor analysis involving differential forms and their Lie transportations.
  • Elasticity: Interaction between strain and stress, which are dual vector-valued forms related by a variational principle.
  • Fluids: Navier–Stokes equation, vorticity formulation, surface waves, Eulerian and Lagrangian methods.

Course Logistics

Getting Started (make sure to follow it in Week 1)

  1. Sign up to Piazza and Gradescope. 
  2. HW0 is due 4/10 (Thu). 

Class Rules

  • Do the assignments individually. Discussion is strongly encouraged, since there are many challenging new mathematical concepts and part of the learning experience is to try out your understanding to your peers.
  • You can look up coding/math questions online. You are also welcomed to post questions (and answer others' questions) on Piazza.
  • Implement your own project. Do not directly download existing code from the internet. You may use existing subroutines if they are only small components of the main system.
  • Usage of ChatGPT/LLM: Treat an AI language tool like ChatGPT as a real colleague/classmate/friend who is not a co-author of your homework/project. You may ask them to proof-read your writing or give you advise to your grammar or even discuss some exciting ideas. But you wouldn't have your colleague or friend to just write paragraphs, code, or math exercises for you. (When discussing ideas, beware that your AI friend usually double down on nonsense. Please use scientific methods to rule out nonsense!)

Grades

There is no quiz or exam. The grades are made of the a few HWs and projects:

Late policy

  • HW late penalty: You have one quota of using a 24 hour extension.