beautypg.com

National Instruments NI MATRIXx Xmath User Manual

Page 78

background image

Chapter 4

Frequency-Weighted Error Reduction

Xmath Model Reduction Module

4-8

ni.com

The left MFD corresponds to the setup of Figure 4-3.

Figure 4-3. C(s) Implemented to Display Left MFD Representation

The setup of Figure 4-2 suggests approximation of:

whereas that of Figure 4-3 suggests approximation of:

In the LQG optimal case, the signal driving K

E

in Figure 4-2 is white noise

(the innovations process); this motivates the possibility of using no
frequency dependent weighting in approximating G(s) [but observe that
after approximating, the signal will no longer be white noise, so that
argument is questionable]. Simple appeal to duality motivates using no
frequency dependent weighting for H(s). These are two of the options
offered by

fracred( )

.

Two more

fracred( )

options depend on examining stability robustness

(the options are duals of one another). From the stability point of view, the
set-up of Figure 4-3 is identical to that of Figure 4-4, with

.

K

r

sI

A

K

E

C

+

(

)

1

K

E

B

P s

( )

+

-

G s

( )

K

r

C

=

sI A

BK

r

+

(

)

1

K

E

H s

( )

K

R

sI A

K

E

C

+

(

)

1

B K

E

=

Pˆ

P I

=