Monday, September 8, 2008

2.8

Main Results
The main idea of this section is numerical approximations of integrals using computers.  The first technique presented is Euler's method.  This method entails calculation x and the velocity of x multiple times over very small steps.  This is the simplest form of numerical integration, and is the least accurate.  The methods increase in order up to the fourth-order Runge-Kutta method which is a good balance of accuracy and non-tediousness.  This method involves finding 4 values and then x(n+1) is a number based on these values.  Avoiding excessively small step-sizes is important because computers round-off at every calculation.

Challenges
My main question from this section is: how do you get a computer to do this for you?  The book mentions writing your own numerical integration routine but I'm sure there is a program already available.  Are we using numerical integration to check answers, or to find solutions when other methods don't work?

Reflections
I like the idea of numerical integration because it brings back fond memories of checking my answers in Calculus using the fnInt function on my calculator.  I have a definite interest in computer science, also, so I am looking forward to class tomorrow and to working on this section.

No comments: