ENGINEERING TRIPOS PART IIB - 2012/2013
Module 4A2 - Computational Fluid Dynamics
|
Leader:
|
Prof P.G. Tucker (pgt23@eng) |
|
Timing:
|
Michaelmas Term
|
|
Prerequisites:
|
3A1 and 3A3 assumed. Pre-module reading about Fortran helpful
|
|
Structure:
|
Coursework with integrated lectures
|
| Assessment: |
Material / Format / Timing / Marks
Progress Check / Brief Report / Week 6 of Michaelmas term / 25%
Coursework / Report / End of Michaelmas term / 75% |
AIMS
This module aims to:
- Provide an introduction to
the field of computational fluid mechanics;
- Help students develop an
understanding of how numerical techniques are devised and analysed with solution of fluid flow problems as the target;
- Provide some experience in
the software engineering skills associated with the implementation of these techniques in practical computer codes;
- Illuminate some of the
difficulties encountered in the numerical solution of fluid flow problems.
LECTURE SYLLABUS
Basic Numerical Concepts (2L including examples), plus demonstrations)
- Finite difference, finite volume, finite element approaches
- Difference scheme and molecules;
- Stability
- Dispersion and Diffusion errors, Cell Re.
- Boundary conditions
Introduction to Advanced Concepts (4L) (Prof. P.G. Tucker)
- Advanced numerical techniques
- Complex modelling - DNS/LES/RANS
- Complex geometries - Geometry definition, mesh generation, parallel computing
- Aerospace CFD in industry lecture
COURSEWORK
Mesh generation (Coursework: approx 2 hours)
- Conversion to Fortran; examples of Fortran programs
- Mesh generation for simplified geometries (eg bend, nozzle, hump, airfoil)
2-D Euler/Time Matching
(Coursework: 5 mini-exercises of about 2-4 hours each, forming a 16 hour mini-project)
- Finite volume discretisation, evaluation of fluxes. (4h)
- Application of boundary conditions. (2h)
- Time Iteration, simple LAX method. (2h)
- Convergence & accuracy testing. (4h)
- Enhancements, e.g. deferred corrections, Adams - Bashforth RK integration, use of energy equation. (4h)
OBJECTIVES
- To be able to formulate
numerical approximations to partial differential equations.
- To be able to write
computer programs for solving the resulting difference equations.
- To understand the
limitations of numerical methods and the compromises between accuracy and
mean time.
- To appreciate the power of
numerical solutions to predict complex flows, including shock waves.
- To develop the critical skills necessary to respond to and audit simulations produced by CFD for complex flow problems.
REFERENCES
Please see the Booklist for Group A Courses for references for this module.
Last updated: September 2012
teaching-office@eng.cam.ac.uk