Computational Fluid Dynamics – my winter quarter adventure

This course may kill me as I know nothing of fluid dynamics, but I am hoping the computational focus will play in my favor. I’m not so interested in macro-scale behavior but more focused on nano-scale applications.

Computational Fluid Dynamics (CFD) is a set of methodologies to solve numerically the governing equations of fluid motion. In the past decades, the development and use of CFD has widely grown in both academia and industry to perform fundamental studies and engineering computations of fluid flows, e.g. for the design of airplanes, turbine blades, jet and rocket engines. This course is an introductory course to CFD covering its fundamentals, as well, few advanced topics.

The students completing this course in good standing are expected to learn: 1. the fundamentals and few advanced topics in CFD; 2. to select and implement numerical schemes for solving model equations for fluid dynamics; 3. to write and execute their own CFD codes (in Fortran or C); 4. to postprocess and analyze CFD results; 5. to write technical reports on CFD results.

2 thoughts to “Computational Fluid Dynamics – my winter quarter adventure”

  1. Hate to be the harbinger of bad news, but industrial CFD (i.e., FEM, FVM) typically assumes that a fluid is a continuum, which does not hold down to molecular (nano) scales. You might have more fun looking into Lattice Boltzmann CFD methods that deal with fluids at the particle (Lagrangian) level. However, practical LB CFD typically assumes packets of particles, so nano-scales are likely to again be troublesome.

    Thought it would be useful to give you an idea what are signing up for.

  2. i am doing some micro-level stuff as well as nano-level stuff. so the continuum assumption is still appropriate for some of what i am doing, but completely appreciate the heads-up. i am interested in multi-scale modelling and so would like to learn methodologies at all the relevant scales. but eventually will have to bust thru and look at particle-level models.

    thanks again

Comments are closed.