Trying to use invfreqs
in [Matlab]
|
Prev: toeplitz matrix
Next: Finding turning points in a curve
From: Madhu Shurpali on 23 Feb 2010 09:23 "Oskar Vivero" <oskar.vivero.nospam(a)gmail.com> wrote in message <hm0mk0$8bv$1(a)fred.mathworks.com>... > "Madhu Shurpali" <madhu.shurpall(a)cvgrp.com> wrote in message <hm0eqn$kp1$1(a)fred.mathworks.com>... > > "Madhu Shurpali" <madhu.shurpall(a)cvgrp.com> wrote in message <hlu2qm$3ve$1(a)fred.mathworks.com>... > > > Sorry for the typo in the previous post. It shoul dbe like the following: > > > > > > "I am unable to understand the phenomena" > > > > Somebody please help.. I am not able to go forward from here. > > While using the system identification toolbox is good advice, there are some things you need to check before that, for example: > > 1) Noise levels, if big your estimates will be biased and hence your response won't look alike to your physical plant. > 2) Why are you picking such model orders in the first place. If you over or under specify your transfer function structure your estimates will also be wrong. In the same vein, and if the noise levels are small there is no reason why invfreq shouldn't work unless your model structure is incorrect. > > I hope this helps Thanks a lot for the help. My results were much better when I specified both the Numeraor and Denominator of order 4. Now this brings me to the question, how do we select a proper model or order for the model. Is there any reference that I can look into to get this information. Thanks Madhu
From: Madhu Shurpali on 23 Feb 2010 09:58 Another observation: When I specified an order of 4 for both the numerator and denominator, I observed that initially the respose was good. But as the simulation time increased, the response kept on diverging - increasing in magnitude. So if my simulation time is 10 sec and I input a sinewave of 5 Hz. Initially I get a gain of about 34 which is expected but after about 5 secs the response starts diverging and the gain is about 100-200. And it keeps on increasing.. Can somebody suggest as to what might be going wrong. Thanks Madhu
From: Oskar Vivero on 23 Feb 2010 11:06
"Madhu Shurpali" <madhu.shurpall(a)cvgrp.com> wrote in message <hm0qds$ea5$1(a)fred.mathworks.com>... > Another observation: > When I specified an order of 4 for both the numerator and denominator, I observed that initially the respose was good. But as the simulation time increased, the response kept on diverging - increasing in magnitude. > > So if my simulation time is 10 sec and I input a sinewave of 5 Hz. Initially I get a gain of about 34 which is expected but after about 5 secs the response starts diverging and the gain is about 100-200. And it keeps on increasing.. > > Can somebody suggest as to what might be going wrong. > > Thanks > Madhu Model order selection is a real problem in the scientific community at this point. There are several criterions you can use (such as Akaike's Information Criterion) but it is still complicated. The divergence you see is because (assuming you are obtaining a denominator similar to the one in your previous post) you have a positive real pole, i.e. your plant is unstable. So again, start with smaller orders and work your way up until you find the best possible fit. |