2011/8/17 Jordi GutiƩrrez Hermoso <jord...@octave.org>:
>
> That being said, what is the system identification toolbox? The
> Mathworks makes it look like it's a combination of Simulink with
> something else.
>

There may be connections with Simulink, but on the whole it doesn't
have anything to do with it. The toolbox is based upon the book,
"System Identification - Theory for the User" By Lennart Ljung. It
uses various methods to get models of a system. Basically, it fits
data to differing kinds of models. The method I use ETFE (Empirical
Transfer Function Estimate) in the end gives a frequency spectrum
corresponding to the system's plant (based upon the given input and
output data). It's different than just doing a FFT on the output data.
But there are methods for Linear Time-Invariant systems, as well as
time-varying and non-linear systems.  Most of the toolbox is centred
around these models and parameter estimation. Using the ETFE I'm not
limited to a specific model since the approach treats the system like
a black box getting a model of what happens based upon input and
output data alone rather than trying to fit to a specific modelling
approach (other than assuming the system is linear).

Documentation:
http://www.mathworks.com/help/toolbox/ident/

And the specific function I use most:
http://www.mathworks.com/help/toolbox/ident/ref/etfe.html

One of the advantages is that I can customize the output frequency
resolution. This is handy when I'm trying to find the natural
frequency since I need good resolution around the damped natural
frequency peaks.

Cheers,

Tim.



-- 
Tim Lahey
PhD Candidate, Systems Design Engineering
University of Waterloo
http://about.me/tjlahey

-- 
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to