Math 621 - Numerical Analysis II

Numerical Methods for
Partial Differential Equations

Fall 2003 - Matthias K. Gobbert

Section 0101 - Schedule Number 3842

This page can be reached via my homepage at

Project Presentations

The project presentations will be held on Monday, November 24, 2003 and Wednesday, November 26, 2003 starting at 05:30 p.m. in MP 401. Please follow the link to the Program for the titles and abstracts.

Grading Information

Scores and grades will be posted here at the end of the semester.

Basic Information

Course Description

Many models for physical processes in nature and in engineering consist of partial differential equations. The models are as varied as reality itself, but often non-linear and often involving systems of partial differential equations. In all but some textbook examples, an analytic solution is impossible. That necessitates the use of numerical methods for partial differential equations, and this area forms a vast field itself and is one of the major driving forces behind research in many other fields like numerical linear algebra, scientific computing, and the development of parallel computers.

Despite their many forms, many equations share certain fundamental mathematical properties and can be classified into the three basic categories of elliptic, parabolic, and hyperbolic partial differential equations. It makes therefore sense to study the mathematical properties and numerical methods for linear prototype equations of each type. Classical examples for the three types are the Poisson equation, the heat equation, and the scalar transport equation, respectively.

This course will provide an overview of the types of equations, their most fundamental mathematical properties, and demonstrate numerical methods for them. Two large classes of methods are finite difference and finite element methods, and we will discuss examples of both methods for each prototype equation. We will use this as the basis for discussing the associated issues of discretizing the time-direction and solving large sparse systems of linear equations efficiently with respect to memory and computing time.

For the finite difference methods, we will write our own code; I suggest Matlab for this purpose because of its ease of programming, but you will probably need to learn additional commands and techniques to get the best resolution and fastest performance. For the computational experiments on the finite element method, we will use FEMLAB, a commercial package based on Matlab and available across campus. It has a sophisticated graphical user interface and is sufficiently powerful to allow the solution and visualization in two and three dimensions.

We will focus simultaneously and equivalently on computational experiments and on rigorous mathematical analysis of the numerical methods considered. Hence, you are expected to have background knowledge equivalent to our first-year graduate courses, in particular Math 620 and Math 630. Additionally, you should have a good foundation in mathematical analysis and be ready to learn more.

Information to Download

This area might be used later to distribute some computer files.

Other Information

UMBC Academic Integrity Policy

By enrolling in this course, each student assumes the responsibilities of an active participant in UMBC's scholarly community in which everyone's academic work and behavior are held to the highest standards of honesty. Cheating, fabrication, plagiarism, and helping others to commit these acts are all forms of academic dishonesty, and they are wrong. Academic misconduct could result in disciplinary action that may include, but is not limited to, suspension or dismissal. To read the full Student Academic Conduct Policy, consult the UMBC Student Handbook, the Faculty Handbook, the UMBC Policies section of the UMBC Directory for undergraduate students, or the Graduate School website for graduate students.

Copyright © 2001-2003 by Matthias K. Gobbert. All Rights Reserved.
This page version 4.0, November 2003.