state space to transfer function conversion vi inconsistency

[grahamwebb]

The SI-estimate-state-space-model.vi outputs the system matrices in a different format to the CD-convert-to-transfer-function.vi input. I can do some transposing and padding of new arrays but this lack of consistency doesn't seem right from NI. Is there a more appropriate vi?

[AdamB]

Hi there,

I'm not very familiar with these vi's.  I assume they are Control Design Toolkit vi's? and System Indentification toolkit?

Which option in the polymorphic vi are you chosing? (the CD-.... has 2 options)

I'll look into it as soon as I have installed the toolkit you mention.

Thanks

AdamB

Message Edited by AdamB on 06-28-2007 04:26 AM

[grahamwebb]

I'm using Labview 8.0 (prof. dev. sys.), the CD convert to transfer function model.vi came from Control Design Toolkit 2.1 and the SI estimate state space model.vi from the System Identification Toolkit 2.0. The contents of the model clusters are slightly different.

I was trying to setup a way of inputting known data points from a bench test (frequency vs gain) to generate a filter with the same freq. response. Can you recommend a way to do this? 

[AdamB]

Hi again,

I found a halfway house but I can't seem to find a vi that does the full conversion.  If you search for "CD Convert Simulation to Control Design.vi"  You will have a vi that outputs the right data structure.

The input is close to the output for the coeficients but not perfect.  I'll keep looking.

AdamB