Lecture 13

IIT Kanpur July 201
20 followers
Feb 2, 2024
[Music]
[Music]
in this lecture we'll be studying how
the Define control inputs going to be
affected on the attitude convergence of
the aircraft we'll find the relationship
between the uh attitude of the aircraft
with respect to the reference signal
attitude of the aircraft with respect to
disturbance as well as the
noise uh then we'll see the combined
effect of the of this control inputs on
the attitude convergence of the aircraft
and also be studying how the controler
going to be plays an important role for
mitigating this disturbance which is
basically unfavorable disturbance to the
system then we'll conclude the
lecture uh we'll continue from the last
lecture in the last lecture we
we had the close loop feedback controls
for the aircraft attitude systems uh let
me redraw the figure again then we'll
proceed so we
had
our
control control transfer function GC
office and the output from the
controller we had U
office and we had summing
point and in the summing Point there's
two input one is MDS which is the
disturbance and the
sum output from this block was going to
the
plant which is GP
office and the output from the plant
was y of s
and we had another summing point we are
having the
reference
input
RS this is we can say yds or we can say
Theta
DS and we had another
input to this summing
point so we had
noise the
system
WS this a negative
feedback
and
the sensor transform function we assume
h of s equal to one this is basically
unit feedback
system
and this is the
input
the uh sensor transform
function and the error was going to the
controller this is e
ofs
now uh in this lecture we'll be finding
the transform
function uh y offs with respect to
disturbance MDS y offs with respect to
R of s the reference signal y of s with
respect to W of s so how the different
transform function we can derive so
let's start so since we are handling the
C system and transform function
generally we come up with for the uh C
single input and single output system so
we can't take all the inputs at the same
time we have to take one at a time so
let y are
R be the
response to the reference
signal
[Music]
signal
with
MDS equal to zero and WS equal to
Z so first we find the TSO function y s
with respect to
RS which
is when we are having Ys with respect to
RS I'm denoting ysb RS y
RS and we are assuming the other input
signal assume to be
zero and in this situation our Clos Loop
trans function can be drawn as this is
the summing
point
where we are
having
RS as a reference
signal
and here we have controller
[Music]
block GC
ofs and since we are not
assuming the
disturbance the
input is going to
the controller
this is my U
ofs this is the GP offs and the
output for this
case I'm assuming y
RS and also we are assuming there is no
noise this is
my Nega negative feedback this is
positive so this is how in this
condition this is how we can draw the
blog diagram uh let me draw this figure
number one this is
figure two now let's find the transer
function y RS with respect to
RS from figure
two we can
write y
RS we can y y RS equal to okay here we
let's denote this is my e
office so it will multiplied GP into GC
into s so
GPS
GCS and Es we can write and also we have
es equal to we can write RS minus y
RS so that we can
write y
RS equal you can
write
GPS
GCS in place of es we can write this
expression or s minus y r
s this you can write and if you multiply
we'll have GP of
s GC of
s RS minus GP of
s GC of
s y
RS
and after rearranging the equations we
can write
y
RS plus GP of
s GC of
s y RS equal to GP
offs GC
offs
RS further we can write we can take y RS
common from the left hand side y
RS
1+ GP of s
GC of
s equal
[Music]
to GP of
s GC ofs and
rs and from this we can
write y
rside by
RS equal to GP of s GC of s/ by 1 + CP
of
s GC of
s so this is
basically the Clos Loop transer function
Clos Loop transer
function which
yelds the
relation
between y
RS
and
rs this is basically the main transform
function because in this transer
function how we going to track
RS so let's denote this equation
number this is equation number
one so first trans function we have
found now we'll go to the second trans
function which is respect
to disturbance the effect of disturbance
on output so now let let now
let y
DS be the
response the
response to the
disturbance
MDS
with so we have to consider one input at
a time so other input will assume to be
0 with
RS = to0 and WS equal
to0 and in this situation let's draw the
figure the Clos Loop diagram this is our
summing
point and this
is
RS which is equal to zero we
assumed
and controller block will be
there this is GC
office and in this case we'll assume
disturbance which is
MDS and this is
us and the sum input sorry some output
going to the
plant which is GP of
s and the output we are denoting here oi
DS and also we are assuming Unity
feedback with WS equal to
Z so this is the close look a
function
when MDS only acting other inputs are
assumed to be
zero and let's denote this figure
number
three from figure
three
we can
write
yds equal to GP of
s into MD of
S
Plus GC of s e of s so again you can
write e of s here e of s you can
write and if
you multiply the factors GP of is
MDS plus GP
ofs GC
ofs yes if you notice
here here basically e of s we can write
minus
yds
because
because R is is assumed to be zero here
because es actually we know
es R of s
minus
yds so here RS is zero so it's only
yds now if you substitute this
expression in this equation we'll
get y of s = to GP of
s
MDS minor GP of s GC of
s yd
ofs after rearranging the equation we
can write
yds plus GP of
s GP of
s GC of
s yds equal to GP of s MD of
s and from this expression we can take
yds
[Music]
common
1+ GP of
s cc
ofs equal
to GP
ofs
MDS and further we can
write
yds by MDS input reference input not
reference disturbance
input and GP
offs
1+ GP ofs GC of
this so this is the second
equation which relates between the
disturbance and output signal so how the
disturbance going to effect on the
output I mean output means the attitude
of the aircraft how it's going to affect
this is the
relation now we'll get the final transer
function with the effect of
noise Omega is so
let y
WS be the
response with the
response with the measurement
noise measurement
noise or
WS so here will assume RS to be zero and
MDS zero so in this condition let's draw
the figure how the figure will be look
like so this is our summing
point this is the reference
signal which is R of s which is assumed
to be zero and this is the error signal
we are producing so this is your is
and
the control is DC
office and
here we can have a suing Point here so
here
MDS to Zer we have already
assumed
and this response is going
to
controller which is zp
ofs and the output from this we are
denoting
ws and here we are
assuming there is a suming
point and
where WS
acting
and this is my feedback
loop this is plus this is plus and this
is a negative fedback
and from this figure we can write from
figure this is let me figure number
four from figure
four we can
write y
WS equal to GP of
s GC of s into es
s but here here actually since we are
assuming measurement noise in the system
and we can write e of s we can
write R of
s minus y
ws+ WS but here we we already assume
that R is to be zero so we can write the
error is basically minus
y
ws+
WS now substituting this expression in
this equation if we substitute this
expression in the previous equation we
can
get equal to we can
write
minus
GPS
GCS y
WS plus
ws and from this we can
write y wsal to minus
CPS
CCS y
WS minus GP of s GC of s
WS if you rewrite the equation rearrange
the equation we can write y WS
S Plus GP of
s GC of
s y
WS = to minus GP of
s GC
ofs
WS if you take y WS common from the left
hand
side we can write
1+ GP of is
GC of
s equal to
minus GP of s gcfs
WS we can write forther y
WS IDE
by
WS equal
to minus GP of
is GC of is divide by 1 + GP of s GC of
s so this is the transer function which
relates between the output and
measurement noise in the system so this
is let me Define this equation number
three if you notice very
carefully in this uh three T of
function that denominator part in all
equation are
same if you see the first equation this
is our denominator
part the second
equation is the
denominator in third
equation also denominator so denominator
part is remain to be same but this is
very very important
observation how we can see the system
with the different inputs disturbance
noise reference
signal okay
now since the system is linear we can
combine the effect so let's
write the combined
response combined
response to
RS
MDS and
WS is the
[Music]
sum is the
sum so we can write let's assume Y is is
the total response we can write y
RS plus
yds plus y
ws and we can further write so y RS
basically we can
write y RS we can write y RS equal to
this whole T of function into RS
similarly y DS we can write the whole
trans of function with
MDS similarly
the Y WS equal to we can write the whole
transform function into
WS so we can let me
write GP of s GC of s / by 1 + GP of s
GC of
s into
RS uh plus GP of
s 1
+ GP of
s GC of s into
MDS and
minus GP of s
CC of
s IDE
1+ CP of
s n GC of s into
ws and let's write this equation
number
four so in this
equation many things we are going to
observe this is very very important
equations our main motivation
is if you see the control objective our
control objective is control
objective y
RS matches
with
RS this is our main objective or we can
say matches with
Theta this is Theta is should match it
with Theta d
or Y of T should matches with y d t this
is our main motivation because R is
basically desert signal or reference
signal the system so this is the main
motivation in control system how we can
come up with our D signal with the
application of
control but if you look in this in
equation in equation
four in the equation four if you notice
carefully GP of as is the
plant in the plant we can't do anything
plant is coming from the natural motion
of the system natural Dynamics so plant
is coming from the
natural
Dynamics so we can't can't play anything
on on this Con function this is the
basically the main plant transer
function so we can't do anything
here and and the other T of function GC
of s is the
controller so controller
basically it can be artificial device or
some human
being some kind of mathematical
formulation we are doing here which is
basically the control Engineers does it
so
here we can say this
is
this is up
to this is up
to
the
control
engineer so we can modify it we can
change that to the defined control
algorithms it is up to us so now if we
can if we can design some control
algorithm
if
if if you can design some control
algorithm where GC
ofs
magnitude is very large because in
control basically there are a lot of
gains will be there we'll be talking
later but let's assume the control
algorithm we designed is very
big let me
write let the
control
be
large so in this
case
if GC of
s
and GP of
s
is the magnitude of is magnitude of this
value is bigger than
one then what you
get then we can can
say the first term in equation four this
is going to be one because this term is
bigger than one this
term then both will be cancel out one
you can ignore here so this term and
this term will cancel out so we can say
in this
situation in this
condition we get we
get y
RS goes
to
Ys but there's a problem okay let's not
not to go to the problem again now so
now if it is very large again let's go
the second
term then we can if it is very large DC
office compared to GP of is so this
term we can say
goes to zero so it means even the
disturbance is there in the
system Ys can track
RS because with this
assumption the disturbance acting on the
system can be
neglected in this condition this is very
very important takeway from this
discussion so first if you can design
our control if you can design our
control algorithm this control algorithm
in such way that this condition will
fulfilled and then we can get let me
write here we can
get from this disc we can write GP of
s GC of s divide by 1 + GP of
s GC of s is goes to one
and GP of
s 1 + GP of s GC of s goes to zero so
this is very I mean good thing that if
you design control in such way that both
this condition can be satisfied so we
can get perfect
tracking but in this
condition there's a one problem in this
condition in with this
condition there's a
problem on the noise so with this
condition we can
write with the
condition with the
condition y
WS is going to be minus WS because if
you look this this equation
our this
[Music]
equation this equation going to be one
what what you have done here if you
notice here this is what you have done
so this with this condition this is the
case this is the problem so if this is
arises we can't get perfect tracking
because WS also in the
equation so this is the compromise we p
as a control system engineer
so this problem can be tackled if you
come up with some frequency analysis
like Nyquist or B plot those thing
will'll be talking later so in because
noise basically noise are basically high
frequency high frequency signals high
frequency frequency
signal so this thing we can tackle while
we'll be designing the control algorithm
based on in the frequency domain so that
using that concept we can tackle this
part as well but this is the main
motivation that in this in this in this
lecture that this is what you have done
here this is if you can do at least we
can get our tracking but this can be
done with with the help of some
navigation or navigation algorithms or
some frequency analysis design based
control we can tackle this problem as
well
and this is the main motivation of
Designing control control algorithm so
that we can get our tracking algorithm
so and we can tracking algorithms we can
get our Des attitude angles with the
help of control algorithm which should
be designed very carefully so that we
can get this kind of objectives let's
stop it here we'll continue from the
next lecture so next lecture we'll be
talking we'll be disc discussing how we
can come up with some typical control
algorithms and how we can get our Des
response in the close loop control
system and we'll be designing all this
algorithm for the aircraft autopilot
design how we can control the
orientation of the body uh of the
aircraft let's stop it here we'll
continue from the next lecture thank you
very
much
[Music]
[Music]
[Music]
