Fundamental Matrix - W4P3

IIT Kanpur July 201
20 followers
Aug 14, 2024
so welcome students and in this video we
are going to talk about something called
a fundamental Matrix okay and by the
name of it you do realize this it is a
very important concept so what is it so
essentially let us consider this system
so consider the system so a variable
coefficient system in Cross system which
is given by X Prime = to 80 * X yes
where where a of T this is a variable
coefficient of course is a continuous en
Crossing Matrix function okay
continuous n cross and Matrix function
here let me put it this way so basically
what I mean is if all the coefficients
of a uh and they are going to be
continuous on whatever interval we are
going to Define this system yes now the
thing is this see we know that if we
choose see if one choose
F of Dy to be 8 * y sorry F of TX to be
a * X TX to be a * X yes and since a is
given to be continuous see a uh here
with respect to T A is discontinuous
right because a only depends on T and F
of TX with respect to just the second
variable X it is C1 right so basically
we can use the picar theorem which we
talked about in the earlier video so if
we choose this then then by beard
existence and uniqueness
beards
existence and
uniqueness and
uniqueness
one uh can find a unique solution right
one can find an unique
solution unique solution and what is the
solution given by let's just call it X
of T okay and the solution is in of
course is in C1 of I yeah C1 of I to RN
satisfying what that some initial data
right
satisfying X at the
point let's say Z is X not okay so
whenever you are given this initial data
you have a solution which passes through
the initial data that is why cards
existence and uniess CLE now what
happens is this see uh if for this
system let's just call this system one
okay now see this x notot is in RN right
yes and you see that now if if one
choose X not okay the vector X not okay
to be let's say EI what is EI which is
nothing but the unit Vector okay so 0 0
and one in the I variable and then 0 0 0
0 yeah let's say this is a element in RN
so basically this is the I entry the
unit i unit Vector if you choose that
then look at this system see uh consider
this system now so consider the
following following
system see what is the system it is
given by F Prime = to 80 * X right 80 *
x and x at the point 0 is
EI okay given this system let's just
call this system as two okay two and I
let's just call it 2 I yeah because this
depends on I right then there exist we
know that a unique solution X let's just
call this solution as x i of T which is
of course is in C1 of I2 RN right
satisfying
[Music]
satisfying 2 I this equation for what I
for I between 1 and N so what I'm trying
to say is this see the thing is this x
not so basically what we are saying is
this this equation one okay that is the
trajectory which a particle is following
yes that's what given and what is x0 at
the point x not that is basically saying
that at the point T equals to 0 so when
time is zero it has to pass through the
Vector X not right so we can choose the
vector to be E1 E2 n right so e e is are
the unit Vector we can choose that and
for each such choice I will get a xite
right I will get a xite now the thing is
this you see let's let us Define this
let us
Define this this
uh F of T let's define this F of T what
is f of t f of T this is what we are
going to call a fundamental Matrix which
is nothing but X1 of T so if you
understand X1 of T is basically a vector
in RN right so you just write down as a
First Column so what is the First Column
of P which is X1 of T what is the second
column of T it is X2 of T okay so what
we are going to do is we are going to
Define this new uh Matrix you understand
what is the definition that the I column
of the fundamental so basically the I
column
column of Ft ft is what we will call it
a fundamental Matrix okay fundamental
Matrix fundamental Matrix
will will
be the XIs is it okay so basically the
XI of T will actually form the I column
is it okay so that is your fee now you
do realize that Fe is are in Crossing
Matrix right and since X1 X2 xn they are
all C1 so basically Fe is also C1 so you
you do not really have to worry about it
yes now okay fine now the thing is there
are some few properties which we really
need to understand
properties first of all if you look at
it you see Fe where is it defined Fe is
defined on I you see for every T from
the interval I you are going to get a
matrix n cross n Matrix right so Fe as a
map yeah it is maap from I to where it's
a a matrix to n cross n okay over R so
that is your Fe is C1 okay it is C1 so
at least continuously differentiable why
it is C1 because you do realize that
every component of it is continuously
differentiable so basically Fe isy
differentiable is it okay right now the
thing is this is the first property
which we know now the second property
this property I am going to write it
down you have to verify it it is very
easy yes it's actually trivial so you
please verify right you see F since it
is C1 what happens to F Prime of T you
can actually show that F Prime of T is
nothing but a * F of D is it okay yeah
this is what Fe does yes and if you
check this part you see
also
also do you realize what f of0 is Right
f of0 is nothing but you see X1 at the
point 0 you see this is X1 the vector X1
at the point 0 X2 at the point 0 right
and xn at the point 0 right that's your
F of 0 so what is that Vector what is F1
at the point 0 you see P1 at the point 0
is nothing but E1 right so the first
Vector so it is 1 0 0 0 and the second
one is 0 1 0 0 you understand so the
last one is all 0 zos everything is zero
and it is one so basically it is nothing
but the identity Matrix is it okay so Fe
hence hence let me put it this way hence
fee satisfies The Matrix fee okay that
satisfies The Matrix differential
equation okay so this is a different
kind of differential equation it's
called a matrix differential
equation differential
equation and what is it given by given
by given
by V Prime of T equals to 80 * VT right
and V at the point 0 is
identity yes now I really do not have to
tell you that since V at the point0 is
identity you can use A's theorem right
yes you can use a theorem and you can
say that the determinant of P of T is
never going to be zero for all T in
I yes see the thing is initially what is
happening is this P at the point 0 is X1
at the point 0 X2 at the point 0 X10 at
the point 0 right see X1 at the point 0
x 2 at the point0 which is nothing but
the identity Matrix so Fe at the point 0
is identity Matrix is it okay now what
happens is of course you I mean the in
the last video we talked about this Abel
formula right what is Abel formula it
says that the Ronan of X is nothing but
Ronan at some point 0 times the
exponential of at right trans of 0 now
the thing is this see here what is
happening is uh you take that arbitary
point x not to be zero here so if you
take that then you do realize that uh
since that
Matrix if you taking the Rong scan that
Rong scan of the identity Matrix is one
so it is basically never going to be
zero okay so in that case determinant of
f of T is non zero which means therefore
therefore
f is non- singular is it
okay so it is basically a non- singular
Matrix Matrix okay what are we using
here we are using Abel's formula
formula okay so you guys have to check
this part out please do that yes okay so
now let's look at some examples okay
what are the fundamental matrices so
let's look at the examples see the thing
is you can easily check that uh if you
define
define F
of sorry t0 let's Define f of t0 which
is given by e to the power a if you
define this thing yes you can see that
this is the fundamental Matrix or let me
call it principal fundamental Matrix I
will explain to you what is principal
why we are calling it principal
fundamental Matrix again this Matrix
also we'll call it a principal
fundamental Matrix okay
principle I'll explain to you why we
call it principle principal
fundamental
Matrix Matrix of what of the
system of the system I think you can
guess this part it is given by X Prime =
to the constant coefficient system X
Prim = to a * X is it okay a is constant
is
constant okay of course n cross n Matrix
a constant so you can check this part
yeah so if you are defining Fe to be
this then Fe will be a fundamental
fundamental Matrix of this system okay
and you do realize that any solution of
this system in that case will look like
f of t0 * X right so that's just any
solution which look like this okay so
now one thing I want to yes principal
fundamental M what is the principal
fundamental you do realize that you see
here this X1 X2 x0 = to e we chose right
e are what e are basically the basis of
R is it okay now the thing is this if we
change the basis of e r from EI to some
other basis right so you choose any n
linear independent vectors and you have
another basis so let's say you know A1
A2 A2 is another basis of RN right it
can have another many different bases
right let's say this is a basis of RN
okay now if you take this equation x
Prime equal to a * x and x at the point
0 is a i let's say yeah you will get a
different set of XI different set of
solutions right those Solutions also you
can put it like this like fee yeah you
can define a different fee and that will
also give you a fundamental Matrix again
the same sort of thing same properties
hold all the same property holds of
course C of Z will be something else but
the thing is doesn't really matter
exactly the exact same so Fe will Prime
T = to 80 * p and all other properties
will be same except that Fe of Z will be
a little bit different it will be a
different Matrix which is by the way
non- singular is it okay and again by AB
formula you can prove that it is a non
singular Matrix so that's not a point
the thing is this you see there can be
many different fundamental matrices for
a given system you understand B based on
the basis which you are choosing
generally speaking if you are choosing
the standard basis which is e what I
chose here then that fundamental Matrix
we'll call it a principal fundamental
Matrix I hope this is clear okay right
so this is uh the thing now the thing is
uh you can also find I mean if you want
yes uh so let's me call this thing as a
and let me write down B what I want you
to do is please check try to find find
out the fundamental Matrix for the so
find the find out the
fundamental fundamental
Matrix mental
Matrix of a the system h of the
system the
system X Prime = to
01 - 1 0 * X is it okay try to find it
out huh see what if we can if you can do
that part yeah please do that okay uh so
that is it right now the thing is this
we are going to prove some a very
important theorem now the theorem says
is this so what is the relation between
these fundamental matrices okay so
theorem so it says that if F of
t f of T is a I am writing it like this
fundamental Matrix FM
okay of the
system of the
system X Prime equal to 0 * X okay so if
you have a homogeneous system and F of T
is given to be the fundamental Matrix of
this homogeneous system okay
then any
constant any
constant non
singular non
singular n cross n
Matrix
Matrix uh then for any then for
any constant non singular Matrix uh
incent Matrix Okay C ft * C ft * C is
also a fundamental Matrix is also a
fundamental Matrix of the same system
okay so let's say what I'm saying is
this if you have one fun Matrix you can
of course produce another one just by
multiplying with a constant and please
don't change the order so it is PT * c
yeah * X of the same system clear also
also we can say that every okay
fundamental
Matrix fundamental
Matrix
Matrix of the system X Prime equal to 80
* X
okay of this system is of the form is of
the form so basically you can't find any
other form of fundamental mistakes
except this of the form ft * C Is it
okay where C is again where C is as
above so basically a non singular n
cross n uh Matrix okay constant Matrix
of course okay so what uh what is the
theorem it it's very simple so
essentially what I told you is this you
see there is a principal fundamental
Matrix and there there other fundamental
Matrix depending on the basis which you
choose right yeah so given a system you
can have like infinitely many
fundamental matrices depending on what
basis you are choosing what is the
principal fundamental Matrix it is
nothing but the Matrix which you get
with respect to so basically the
solution which you are going to get with
if you are choosing the the unit
standard unit uh basis elements okay so
e E1 E2 EA now the thing is this what
what this theorem says that what about
the fundamental matx is there any
relation between any two fundamental
Matrix what it is saying is this it
doesn't really matter what you are going
to choose your basis the fundamental
Matrix essentially will look like if one
is given to be Fe another one will look
like ft * a constant c yeah and it's
also true that every fundamental Matrix
will look like this so you can't go on
showing something else is a fundamental
Matrix okay right so let's see okay so
first of all let's say if FT is a
fundamental Matrix we have to show that
ft * C is also a fundamental is very
easy okay now since since f is a
fundamental Matrix what did I tell you
this is true right F Prime of T = to 8 *
F of T now this is quite easy to uh
check as I told you of course yes please
do that so this is one let's just call
that one okay now you see I have to show
ft * C is also a fundamental Matrix so
let's look
now you see
f * C let's just look at the derivative
of this what is it constant C C is a
constant Matrix right so the derivative
of that is basically nothing so it is
essentially P Prime of T acting at C
right again F Prime of T we know it is a
* ft right ft acting at C yes so you see
by matrix we know that associativity
property holds right so you can write it
as a * ft * C okay now f * C see see f *
C Prime is nothing but 8 * ft * C yes
therefore f t and ft * C are both
Solutions okay therefore F of
T and ft * C are both Solutions are both
Solutions of the
Matrix
of this equation uh you see the The
Matrix value differential equation okay
so F Prime equal to *
Fe okay satisfies this so This Is The
Matrix differential equation please
remember yes the fundamental matrixes
they satisfies The Matrix differential
equation okay right okay so now you see
the thing is you have to also show that
since the important thing is how do you
check something is a fundamental first
of all you check s satisfy the Matrix
differential equation and the next you
check is what about the determinant if
it is non- singular or not and you see
see since since determinant of P of T
this is non zero right because F of T is
given to be fundamental Matrix therefore
determinant of Ft * C is nothing but
determinant of Ft * determinant of c and
since determinant of C is non Zer
because C is given to be non- singular
you see non- singular constant Matrix
okay so this is also going to be non
zero so it's trivial right so you can
actually see that deter PT * C is also a
fundamental Matrix clear very simple now
we have to show the part so what is the
converse so let's say Fe of T um any two
fundamental Matrix is given to you yes
we have to show that one is a constant
multiple of another so basically V1 and
P2 are two fundamental matrices you have
to show that V2 is nothing but V1 * C
yes okay so conversely let's do that
conversely we have to
show that
let F1 of T and and F2 of T and F2 of
T B2 fundamental matrices
B2 fundamental
matrices
Matrix
solution
okay of the system X Prime = to 80 *
X okay now the thing is this see if this
is the case
yeah just look at this
thing Fe and Fe inverse both are non-
singular right so I can talk about the
inverses yeah so
Define this new Matrix C of T which is
nothing but
F2 inverse of T time F1 of T let's just
call that okay now if you do that see
what is F1 of T in this case F1 of T is
nothing but F2 of T * CT is it okay yes
and if you take the you see F1 and F2
let's say C F1 and F2 are both C1 map C1
Matrix functions right so C is also C1
right so we can take the derivative here
now differentiating what we have
differentiating one
has one has differentiating this part
okay I have F1 Prime of T right F1 Prime
of t
is nothing but F2 Prime of T *
CT plus F2 of T C Prime of
T right now you see what is F1 Prime of
T F1 Prime of T is nothing but a of T *
F1 of T yes and what is F2 Prime of T it
is a of T * F2 of t * CT yeah plus F2 of
T * C Prime of T yes now you see F1 is
sorry F1 is F2 * CT F1 is F2 * CT this
is F2 * CT so basically this part and
this part we can just cancel it out so
basically you can take the one Matrix to
be on the other side and that will
actually give you zero now essentially
what you have is this F2 of T C Prime of
T is nothing but the zero Matrix yes F2
is what F2 is a fundamental Matrix so
that's a non- singular Matrix if you
take F2 inverse on the both sides
essentially that will give you C2 T is
zero which basically says that CT is
constant is it okay so if CT is constant
you do realize that F1 will be F2 * a
constant Matrix C okay so with this I am
going to end this video
