last modified 23/02/2006

Coursecode: wb1428
Coursename: Computational Fluid Dynamics

ECTS creditpoints: 3

Faculty of 3mE

Lecturer(s): Pourquie, dr.ir. M.J.B.M.

Tel.: 015-27 87979 / 82997

Catalog data:
finite volume method, convection-diffusion equation, stability of schemes, conservation laws for flow problems,steady flow, time-dependent flow, turbulence models, turbulent flow, boundary conditions.

Courseyear: MSc 1st year
Semester: 2A / 2B
Hours p/w: 2
Other hours: 1
Assessment: thesis
Assessm.period(s):
(see academic calendar)

Prerequisites: wb1321, wi2021tu

Follow up:  

Detailed description of topics:

The lecture series Computational Fluid Dynamics (CFD) will help students to use commercial CFD packages properly. The basis is the commercial CFD package Fluent, which is widely used at the TU-Delft and widespread in the industry. Important aspects treated in the course are:

Numerical methods for advection-diffusion equations, as used in the commercial packages. Grid generation and effects of the grid on the quality of the solution.

Numerical boundary conditions. Representation of wall boundaries, inflow boundaries, far field boundaries. Aspects of turbulence models and boundary conditions for turbulent flows.

Pre- and postprocessing in commercial packages.  Solver structure of commercial packages, user adaptable parameters in the flow solver.

During the lectures exercises are given to get acquainted with the software package.

Course material:
J.H. Ferziger and M. Peric, Computational methods for Fluid Dynamics, Springer Verlag.

References from literature:

  • C. Hirsch, Numerical computation of internal and external flows, Volume I Fundamentals of numerical discretization, Volume II Computational methods for inviscid and viscous flows, Chicester, Wiley & Sons, 1988, 1990

  • C.A.J. Fletcher, Computational techniques for Fluid Dynamics, Volume I Fundamental and general techniques, Volume II Specific techniques for different flow categories, Berlin, Springer, 2-nd ed. 1991.

Remarks (specific information about assesment, entry requirements, etc.):

Goals:
The student must be able to:

  1. describe the two most popular methods in commercial CFD, finite differences and finite volumes

  2. solve simple demonstrative problems in fluid flow and heat transfer by programming them in Matlab, using finite differences and finite volumes

  3. recognize the effects of numerical methods on the solution, such as numerical diffusion and numerical dispersion and to explain how to make these effects smaller

  4. recognize numerical instability, to list several ways to avoid it and to analyze stability of simple methods analytically

  5. solve fluid flow and heat transfer problems with the commercial CFD package Fluent, which includes the following:

  • make the geometry in the preprocessor

  • choose appropriate boundary conditions

  • correctly apply wall boundaries, inflow boundaries, outflow boundaries, far field boundaries

  • adapt the geometry to properly include boundary conditions

  • make an appropriate grid, taking into account grid cell quality and grid point density

  • run the solver for the problem

  • choose appropriate flow related quantities to monitor convergence of the solver

  • visualize the results, obtain relevant quantities such as forces on objects and heat flux through surfaces

  • interpret the results and recognize where the geometry and the grid have to be improved

  • find out or argue whether grid-refinement is necessary

Computer use:
Practical exercises with simple examples in order to check convergence, stability, choice of step-length.

Laboratory project(s):
Practical exercises with a commercial code (FLUENT).

Design content:
The design of a correct discretization is part of the practical work.

Percentage of design: 25 %