Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Feedback Control System Lec 16

Feedback Control System Lec 16

Ratings: (0)|Views: 10 |Likes:
Published by Ulemj

More info:

Published by: Ulemj on Jul 17, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

07/17/2012

pdf

text

original

 
Topic
 
#16
 
16.30/31
 
Feedback
 
Control
 
Systems
 
 
Add
 
reference
 
inputs
 
for
 
the
 
DOFB
 
case
 
 
Reading:
 
FPE
 
7.8,
 
7.9
 
 
Fall
 
2010
16.30/31
 
16–2
 
Reference
 
Input
 
- II
On
 
page
 
15-6,
 
compensator
 
implemented
 
with
 
reference
 
command
 
by
 
changing
 
to
 
feedback
 
on
 
e
(
t
) =
 
r
(
t
)
 
 
y
(
t
)
rather
 
than
 
y
(
t
)
r
 
G
c
(
s
)
 
G
 p
(
s
)
 
e u
 
y
 
 
So
 
u
(
t
) =
 
G
c
(
s
)
e
(
t
) =
 
G
c
(
s
)(
r
(
t
)
 
 
y
(
t
))
 
 
Intuitively
 
appealing
 
because
 
same
 
approach
 
used
 
for
 
classical
 
control,
 
but
 
it
 
turns
 
out
 
not
 
to
 
be
 
the
 
best.
 
Can
 
improve
 
the
 
implementation
 
by
 
using
 
a
 
more
 
general
 
form:
 
˙
x
c
(
t
)
 
=
 
A
c
x
c
(
t
)
 
+
 
B
c
y
(
t
)
 
+
 
G
r
(
t
)
 
u
(
t
)
 
=
 
c
x
c
(
t
)
 
+
 
r
(
t
)
 
 
Now
 
explicitly
 
have
 
two
 
inputs
 
to
 
controller
 
(
y
(
t
)
 
and
 
r
(
t
)
)
 
 
 
performs
 
the
 
same
 
role
 
that
 
we
 
used
 
it
 
for
 
previously.
 
 
Introduce
 
G
 
as
 
an
 
extra
 
degree
 
of 
 
freedom
 
in
 
the
 
problem.
 
Turns
 
out
 
that
 
setting
 
G
 
=
 
B
 
is
 
a
 
particularly
 
good
 
choice.
 
 
Following
 
presents
 
some
 
observations
 
on
 
the
 
impact
 
of 
 
G
 
November
 
2,
 
2010
 
 
 
 
 
Fall
 
2010
16.30/31
 
16–3
 
First:
 
this
 
generalization
 
does
 
not
 
change
 
the
 
closed-loop
 
poles
 
of 
 
the
 
system,
 
regardless
 
of 
 
the
 
selection
 
of 
 
G
 
and
 
,
 
since
 
x
˙(
t
) =
 
A
x
(
t
) +
 
B
u
(
t
)
 
,
 
y
(
t
) =
 
x
(
t
)
 
x
˙
c
(
t
) =
 
A
c
x
c
(
t
) +
 
B
c
y
(
t
) +
 
G
r
(
t
)
 
u
(
t
) =
 
c
x
c
(
t
) +
 
r
(
t
)
 
 
 
 
x
˙(
t
)
A
 
B
c
 
x
(
t
)
 
B
 
 
x
˙
c
(
t
)=
 
B
c
 
A
c
 
x
c
(
t
)+
 
G
 
r
(
t
)
 
 
x
(
t
)
 
y
(
t
) =
 
 
0
 
x
c
(
t
)
 
 
So
 
the
 
closed-loop
 
poles
 
are
 
the
 
eigenvalues
 
of 
 
A
 
B
c
 
B
c
 
A
c
 
(same
 
as
 
15–7
 
except
 
“–”
 
in
 
a
 
different
 
place,
 
gives
 
same
 
closed-loop
 
eigenvalues)
 
regardless
 
of 
 
the
 
choice
 
of 
 
G
 
and
 
 
 
G
 
and
 
 
impact
 
the
 
forward
 
path,
 
not
 
the
 
feedback
 
path
 
Second:
 
if 
 
 
= 0
 
and
 
G
 
=
 
L
 
=
 
B
c
,
 
then
 
we
 
recover
 
the
 
original
 
implementation,
 
on
 
15–6
 
since
 
the
 
controller
 
reduces
 
to:
 
x
˙
c
(
t
) =
 
A
c
x
c
(
t
) +
 
B
c
(
y
(
t
)
 
 
r
(
t
))
 
=
 
A
c
x
c
(
t
) +
 
B
c
(
e
(
t
))
 
u
(
t
) =
 
c
x
c
(
t
)
 
 
With
 
G
c
(
s
) =
 
c
(
sI 
 
 
A
c
)
1
B
c
,
 
then
 
this
 
compensator
 
can
 
be
 
written
 
as
 
u
(
t
) =
 
G
c
(
s
)
e
(
t
)
 
as
 
before
 
(since
 
the
 
negative
 
signs
 
cancel).
 
November
 
2,
 
2010
 

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->