Choice of F_Q transfer function in Q-design controller

Translate This Thread From English to

Threaded View


Given a plant's transfer function, G_o(s), how would one be able to
choose the choice of a controller using Q-design method, where Q(s) =
F_(s)*G_o^(s), where G_o^(s) is the inverted signal of the
plant? More specifically, how would one justify whether to use first,
second, or third-order model for F_(s)? Thanks in advance

Re: Choice of F_Q transfer function in Q-design controller



ssylee wrote:

Practically or academically?  (Where have I used that before?)

I'm not sure what you mean by the order of the model, but I assume that
you're constraining the behavior of your F_(s) to keep the plant
drive within bounds -- so do some analysis on the plant transfer
function, or try things out in simulation, and see what works best.  You
want the simplest system that will still attain your goals.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Re: Choice of F_Q transfer function in Q-design controller




It isn't clear what you are really trying to do?  One could do many
things.   if the controller is 1/G_o(s) then F(s) could be a low
pass filter to reduce output noise.  Why not simply make F((s) a
constant to increase the gain and get the closed loop control closer
to unity?

You need to give a precise example as the question is vague.
Why bother with Q design in the first place is a better question?

Peter Nachtwey

Site Timeline