Math 151B,  Section 1

Applied Numerical Methods

Spring2002@UCLA

Most of the numerical methods used to solve problems in engineering and the physical sciences are synthesized from a "core" set of numerical procedures. These procedures concern the solution of linear and non-linear equations, numerical integration, numerical differentiation, etc. The goal of 151A/B is to familiarize students with the algorithms that are used to implement these "core" numerical procedures. We'll be covering the mathematical foundation and properties of these algorithms as well as implementation details. By the end of the quarter the students should be competent at choosing and implementing any of the "core" procedures we've covered, as well as be competent at generalizing their knowledge to address issues involving the procedures that we didn't cover.   For the computer implementations, we'll be using Matlab. Previous familiarity with this package is not required for enrollment. Math 151B:  Numerical solution of differential equations, two-point boundary value problems, iterative solution of linear equations, topics in approximation theory.  

Chapter 5. Initial Value Problems for ODEs (4 weeks)
5.1   Basic Theory of Initial value problems
5.2   Euler's method
5.3   Higher order Taylor methods
5.4   Runge-Kutta Methods 
5.5   Error Control
5.6    Multi-step Methods
5.7    Adaptive methods
5.8    Extrapolation methods
5.9    Higher order equations and systems
5.10  Stability
5.11  Stiff equations

Chapter 7.  Iterative Techaniques in Matrix Algebra (1.5 weeks)
7.1  Norms of Vecors and Matrices
7.3 Iterative techaniques for solving linear systems;
7.4 Error estimates for iterative methods.

Chapter 8  Approximate Theory (2 weeks) 
8.1   Discrete least squares approximation 
8.2  Orthogonal polynomial
8.5  Trigonometric polynomial approximation
8.6  Fast Fourier Transforms

Chapter 11. Boundary Value Problems for ODEs (1.5 weeks)
11.1  Shooting method for linear problems
11.2  Shooting method for nonlinear problems
11.3  Finite difference methods for linear problems
11.4  Finite difference methods for nonlinear problems