| Module Identifier |
PH36010 |
| Module Title |
NUMERICAL METHODS |
| Academic Year |
2001/2002 |
| Co-ordinator |
Dr Geraint Thomas |
| Semester |
Semester 2 |
| Other staff |
Dr David Langstaff |
| Pre-Requisite |
PH26010 , PH27010 , PH23010 , PH21010 |
| Course delivery |
Lecture | 12 lectures |
| |
Seminars / Tutorials | 21 Hours 5 workshops (2 hours each); project lasting 11 hours |
| Assessment |
Course work | 4 Assignments Coursework Deadlines (by week of Semester):
Assignment 1 Week 3
Assignment 2 Week 6
Assignment 3 Week 8
Assignment 4 Week 11 | 100% |
Module description
Computational physics provides an alterative approach for the solution of practical and theoretical problems. Solutions intractable by analytical techniques may be evaluated using numerical techniques or, alternatively, numerical simulation may allow lthe influence of a range of variables to be investigated without recourse to extensive experiments. In the present course, basic techniques of numerical analysis will be introduced, including interpolation, functions, roots and integration. The module will also introduce approaches for the solution of ordinary differential equations and Fourier transforms as well as finite element techniques for solving partial differential equations. A basic knowledge of the FORTRAN programming language is requried, but the module will concentrate on applying existing routines rather than programme development and significant emphasis will be placed on the physical relevance of the sample problems and their solutions.
Learning outcomes
After taking this module students should be able to:
-
demonstrate a familiarity with various techniques for scientific computing and analysis
-
implement and modify given numerical subroutines so as to perform the relevant analysis
-
develop simple numerical analysis codes, based on the governing laws and equations
Additional learning activities
In addition to formal lectures on basic techniques, the students will have significant opportunities to investigate and implement numerical analysis methods on personal computers.
Outline syllabus
FORTRAN: revision lecture
Linear interpolation and extrapolation
Roots of equations
Numerical Integration
Fourier Analysis
Ordinary Differential Equations: Runge-Kutta
Introduction to the solution Partial Differential Equation: Finite Difference techniques.
Each of the above will be illustrated by reference to appropriate topics in Physics
Reading Lists
Books
** Recommended Text
Paul L.DeVries.
A First Course in Computation Physics. John Wiley ISBN 0471548693
** Supplementary Text
William H. Press, et al..
Numerical Recipes in FORTRAN. Cambridge University Press ISBN 052143064X