MAST90026: Computational Differential Equations

4x minor homeworks (worth 5% each), 3x major assignments (worth 20% each) and an oral presentation (worth 20%).

This subject is an introduction to computational methods for solving differential equations that have no exact mathematical solution - which is most of them. The lectures are held in a computer lab but mostly cover the theory side of the course, with a few short demonstrations using Matlab. The assignments are mostly very practical, giving you specific problems to solve that will require you to code up the methods described in class and examine their performance (accuracy and speed).

I found it to be a really enjoyable subject. There is no exam, but the major assignments are very long (mine were approximately 30 pages each, most of which was plots and printouts of Matlab code). The oral presentation is something you'll either love or hate: you get to explain a current research topic in the area to your classmates. It's suggested that to do well you should also explain real-world examples of where your topic is used and why you'd want to use it instead of the methods taught in the class.

There is no formal prerequisite, but if you don't have any programming experience, you're going to struggle. The numerical part of the unit MAST30028 Numerical and Symbolic Mathematics is assumed knowledge, so if you haven't done that before (I hadn't) you'll have a bit of catching up to do in the first few weeks of semester. In terms of required mathematical background, you'll find yourself using your linear algebra skills far more than any methods you would have seen before for solving DEs.

No.

A/Prof Steven Carnie.

5/5

There are three textbooks listed as 'references' in the course outline. The most useful of these is "Finite Difference Methods for Ordinary and Partial Differential Equations" by R. Leveque (2007), and is downloadable from the University library web site.

3 hours per week, combined lecture/computer labs.

2014, Semester 1.

H1