| Code |
8492
|
| Year |
2
|
| Semester |
S2
|
| ECTS Credits |
6
|
| Workload |
TP(60H)
|
| Scientific area |
Mathematics
|
|
Entry requirements |
The student should already be able to:
- calculate derivatives of an actual function;
- operate with matrices;
- use the concept of linear combination in polynomial vector spaces;
- efficiently use the calculator as an auxiliary calculation tool.
|
|
Mode of delivery |
Classroom
|
|
Learning outcomes |
The general objective of this course is the study of efficient and stable numerical methods for solving certain mathematical problems. The study of each numerical method includes the analytic deduction of the formulae used, the description in algorithmic language and the presentation of techniques to estimate the solution error.
This objective is realized by the transmission of the following competences:
a) Analyse errors and determine error propagation;
b) Calculate roots and extreme values of a non-linear function;
c) Solve systems of linear and non-linear equations;
d) Interpolate and approximate random data sets by polynomial functions;
e) Differentiate and integrate functions numerically;
f) Solve differential equations and systems of differential equations numerically.
At the end the student should be able to: Identify possible methods to solve mathematical model of an engineering problem, choose the most appropriate one, implement it in MATLAB and criticize the results.
|
|
Syllabus |
1. Introduction to numerical computing: Floating point computing; Approximation of real functions; Conditioning of a problem e stability of a numerical method. Numerical differentiation.
2. Systems of linear equations: Direct methods and numerical instability; Jacobi and Gauss-Seidel iterative methods.
3. Nonlinear equations: bisection method, fixed point method and Newton's method.
4. Polynomial interpolation: Lagrange and Newton formulas and interpolation by segmented polynomials.
5. Numerical integration: Newton-Cotes and Gauss rules.
6. Numerical methods for differential equations and systems of differential equations: Methods based on the Taylor series and Runge-Kutta methods; consistency, stability and convergence.
|
|
Main Bibliography |
A - “Numerical Analysis”, R. L. Burden e J. D. Faires, 2011, Brooks/Cole. Cengage Learning.
B - “Métodos Numéricos”, M. R. Valença, 1993, Livraria Minho.
- “Introduction to MATLAB with Numerical Preliminaries", A. Stanoyevitch, 2005, John Wiley & Sons.
- “GNU Octave – Beginner’s Guide”, J. S. Hansen, 2011, Packt.
- “Cálculo científico com MATLAB e Octave”, A. Quarteroni e F. Saleri, 2007, Springer-Verlag.
|
|
Language |
Portuguese. Tutorial support is available in English.
|