MIT-OCW-Transcripts / -0_vZ4t-q0I.txt
olimiemma's picture
Add files using upload-large-folder tool
7f94e4c verified
Raw History Blame Contribute Delete
32.1 kB
align:start position:0%
hi everyone welcome back so today I'd
align:start position:0%
hi everyone welcome back so today I'd
align:start position:0%
hi everyone welcome back so today I'd
like to take a look at solving a linear
align:start position:0%
like to take a look at solving a linear
align:start position:0%
like to take a look at solving a linear
o de but with a periodic input and
align:start position:0%
o de but with a periodic input and
align:start position:0%
o de but with a periodic input and
specifically we're asked to find one
align:start position:0%
specifically we're asked to find one
align:start position:0%
specifically we're asked to find one
solution that a particular solution
align:start position:0%
solution that a particular solution
align:start position:0%
solution that a particular solution
which is also the periodic solution of
align:start position:0%
which is also the periodic solution of
align:start position:0%
which is also the periodic solution of
the differential equation X dot plus two
align:start position:0%
the differential equation X dot plus two
align:start position:0%
the differential equation X dot plus two
x dot plus 4x equals
align:start position:0%
x dot plus 4x equals
align:start position:0%
x dot plus 4x equals
the square wave function so the square
align:start position:0%
the square wave function so the square
align:start position:0%
the square wave function so the square
wave function is periodic function with
align:start position:0%
wave function is periodic function with
align:start position:0%
wave function is periodic function with
period 2pi it's defined as minus 1 and 1
align:start position:0%
period 2pi it's defined as minus 1 and 1
align:start position:0%
period 2pi it's defined as minus 1 and 1
on the intervals minus prior to 0 and 0
align:start position:0%
on the intervals minus prior to 0 and 0
align:start position:0%
on the intervals minus prior to 0 and 0
to PI and we know that the square wave
align:start position:0%
to PI and we know that the square wave
align:start position:0%
to PI and we know that the square wave
function has the following Fourier
align:start position:0%
function has the following Fourier
align:start position:0%
function has the following Fourier
series
align:start position:0%
series
align:start position:0%
series
so let you think about this problem for
align:start position:0%
so let you think about this problem for
align:start position:0%
so let you think about this problem for
a moment and I'll come back in a second
align:start position:0%
align:start position:0%
hi everyone welcome back
align:start position:0%
hi everyone welcome back
align:start position:0%
hi everyone welcome back
okay so the reason we've been studying
align:start position:0%
okay so the reason we've been studying
align:start position:0%
okay so the reason we've been studying
Fourier series is to essentially solve
align:start position:0%
Fourier series is to essentially solve
align:start position:0%
Fourier series is to essentially solve
differential equations with complicated
align:start position:0%
differential equations with complicated
align:start position:0%
differential equations with complicated
forcing inputs on the right hand side
align:start position:0%
forcing inputs on the right hand side
align:start position:0%
forcing inputs on the right hand side
which are periodic and the reason we've
align:start position:0%
which are periodic and the reason we've
align:start position:0%
which are periodic and the reason we've
been studying Fourier series is because
align:start position:0%
been studying Fourier series is because
align:start position:0%
been studying Fourier series is because
we know that differential equations with
align:start position:0%
we know that differential equations with
align:start position:0%
we know that differential equations with
sines and cosines as forcing terms on
align:start position:0%
sines and cosines as forcing terms on
align:start position:0%
sines and cosines as forcing terms on
the right-hand side are relatively easy
align:start position:0%
the right-hand side are relatively easy
align:start position:0%
the right-hand side are relatively easy
to solve and
align:start position:0%
to solve and
align:start position:0%
to solve and
we want to be able to solve the same
align:start position:0%
we want to be able to solve the same
align:start position:0%
we want to be able to solve the same
differential equation with a more
align:start position:0%
differential equation with a more
align:start position:0%
differential equation with a more
complicated periodic function on the
align:start position:0%
complicated periodic function on the
align:start position:0%
complicated periodic function on the
right hand side so the general approach
align:start position:0%
right hand side so the general approach
align:start position:0%
right hand side so the general approach
is to first
align:start position:0%
align:start position:0%
do you compose
align:start position:0%
align:start position:0%
the right-hand side
align:start position:0%
align:start position:0%
into a Fourier series
align:start position:0%
align:start position:0%
okay and this step is essentially
align:start position:0%
okay and this step is essentially
align:start position:0%
okay and this step is essentially
already done for us we're told what the
align:start position:0%
already done for us we're told what the
align:start position:0%
already done for us we're told what the
Fourier series is
align:start position:0%
Fourier series is
align:start position:0%
Fourier series is
and what we do is we solve
align:start position:0%
and what we do is we solve
align:start position:0%
and what we do is we solve
the Eau de
align:start position:0%
the Eau de
align:start position:0%
the Eau de
X dot dot plus 2 X dot plus 4x and
align:start position:0%
X dot dot plus 2 X dot plus 4x and
align:start position:0%
X dot dot plus 2 X dot plus 4x and
I'm just going to take one
align:start position:0%
I'm just going to take one
align:start position:0%
I'm just going to take one
term of the Fourier series sine NT
align:start position:0%
align:start position:0%
so if we know the right hand side is a
align:start position:0%
so if we know the right hand side is a
align:start position:0%
so if we know the right hand side is a
sum of a whole bunch of signs what we're
align:start position:0%
sum of a whole bunch of signs what we're
align:start position:0%
sum of a whole bunch of signs what we're
going to do is we're going to solve for
align:start position:0%
going to do is we're going to solve for
align:start position:0%
going to do is we're going to solve for
any specific one of those signs so we
align:start position:0%
any specific one of those signs so we
align:start position:0%
any specific one of those signs so we
want to solve this differential equation
align:start position:0%
want to solve this differential equation
align:start position:0%
want to solve this differential equation
and
align:start position:0%
and
align:start position:0%
and
then we use
align:start position:0%
align:start position:0%
superposition
align:start position:0%
align:start position:0%
so if we know what the solution is to 1
align:start position:0%
so if we know what the solution is to 1
align:start position:0%
so if we know what the solution is to 1
sine NT and we know that the right hand
align:start position:0%
sine NT and we know that the right hand
align:start position:0%
sine NT and we know that the right hand
side forcing is a sum of many sign
align:start position:0%
side forcing is a sum of many sign
align:start position:0%
side forcing is a sum of many sign
entities with appropriate weight factors
align:start position:0%
entities with appropriate weight factors
align:start position:0%
entities with appropriate weight factors
then we can use superposition to
align:start position:0%
then we can use superposition to
align:start position:0%
then we can use superposition to
construct
align:start position:0%
align:start position:0%
a
align:start position:0%
align:start position:0%
final solution
align:start position:0%
align:start position:0%
okay so this is the method to attack
align:start position:0%
okay so this is the method to attack
align:start position:0%
okay so this is the method to attack
this problem
align:start position:0%
this problem
align:start position:0%
this problem
alright so we've already done step 1 or
align:start position:0%
alright so we've already done step 1 or
align:start position:0%
alright so we've already done step 1 or
we were given step 1 and we wanted to
align:start position:0%
we were given step 1 and we wanted to
align:start position:0%
we were given step 1 and we wanted to
solve step 2 and we can solve this just
align:start position:0%
solve step 2 and we can solve this just
align:start position:0%
solve step 2 and we can solve this just
using the exponential response formula
align:start position:0%
using the exponential response formula
align:start position:0%
using the exponential response formula
so what we're going to do is we're just
align:start position:0%
so what we're going to do is we're just
align:start position:0%
so what we're going to do is we're just
going to
align:start position:0%
going to
align:start position:0%
going to
complexity the right hand side
align:start position:0%
align:start position:0%
so if I want to solve the differential
align:start position:0%
so if I want to solve the differential
align:start position:0%
so if I want to solve the differential
equation x + 2 X dot plus 4x equals sign
align:start position:0%
equation x + 2 X dot plus 4x equals sign
align:start position:0%
equation x + 2 X dot plus 4x equals sign
in T I'm going to denote the solution
align:start position:0%
in T I'm going to denote the solution
align:start position:0%
in T I'm going to denote the solution
with a subscript n and it's going to be
align:start position:0%
with a subscript n and it's going to be
align:start position:0%
with a subscript n and it's going to be
the imaginary part of
align:start position:0%
the imaginary part of
align:start position:0%
the imaginary part of
and I'm just using the exponential
align:start position:0%
and I'm just using the exponential
align:start position:0%
and I'm just using the exponential
response formula 1 over the
align:start position:0%
response formula 1 over the
align:start position:0%
response formula 1 over the
characteristic polynomial evaluated at I
align:start position:0%
align:start position:0%
ne2 the int
align:start position:0%
ne2 the int
align:start position:0%
ne2 the int
you might ask how did I get that
align:start position:0%
align:start position:0%
well I basically just took the
align:start position:0%
well I basically just took the
align:start position:0%
well I basically just took the
complexity aight formula or sorry the
align:start position:0%
complexity aight formula or sorry the
align:start position:0%
complexity aight formula or sorry the
complex if I'd equation
align:start position:0%
align:start position:0%
and noted that sine NT was the imaginary
align:start position:0%
and noted that sine NT was the imaginary
align:start position:0%
and noted that sine NT was the imaginary
part of e to the int and then use the
align:start position:0%
part of e to the int and then use the
align:start position:0%
part of e to the int and then use the
exponential response formula and I want
align:start position:0%
exponential response formula and I want
align:start position:0%
exponential response formula and I want
to take the imaginary part at the end of
align:start position:0%
to take the imaginary part at the end of
align:start position:0%
to take the imaginary part at the end of
the day
align:start position:0%
align:start position:0%
so what is a the characteristic
align:start position:0%
so what is a the characteristic
align:start position:0%
so what is a the characteristic
polynomial in this case it's s squared
align:start position:0%
polynomial in this case it's s squared
align:start position:0%
polynomial in this case it's s squared
plus 2 s plus 4
align:start position:0%
plus 2 s plus 4
align:start position:0%
plus 2 s plus 4
which means that P of I n
align:start position:0%
which means that P of I n
align:start position:0%
which means that P of I n
is going to be negative n squared plus 4
align:start position:0%
is going to be negative n squared plus 4
align:start position:0%
is going to be negative n squared plus 4
so I'll just group the real terms
align:start position:0%
so I'll just group the real terms
align:start position:0%
so I'll just group the real terms
together plus 2i n
align:start position:0%
together plus 2i n
align:start position:0%
together plus 2i n
so the 2i n comes from the 2's term
align:start position:0%
so the 2i n comes from the 2's term
align:start position:0%
so the 2i n comes from the 2's term
and
align:start position:0%
and
align:start position:0%
and
then X of n is going to be the imaginary
align:start position:0%
then X of n is going to be the imaginary
align:start position:0%
then X of n is going to be the imaginary
part of
align:start position:0%
align:start position:0%
1 over 4 minus n squared
align:start position:0%
1 over 4 minus n squared
align:start position:0%
1 over 4 minus n squared
plus 2i
align:start position:0%
plus 2i
align:start position:0%
plus 2i
ne2 the int
align:start position:0%
ne2 the int
align:start position:0%
ne2 the int
and
align:start position:0%
and
align:start position:0%
and
I'm going to use the amplify
align:start position:0%
I'm going to use the amplify
align:start position:0%
I'm going to use the amplify
the amplitude phase form to convert this
align:start position:0%
the amplitude phase form to convert this
align:start position:0%
the amplitude phase form to convert this
or to convert this Cartesian complex
align:start position:0%
or to convert this Cartesian complex
align:start position:0%
or to convert this Cartesian complex
number
align:start position:0%
number
align:start position:0%
number
sorry I'm going to convert this
align:start position:0%
sorry I'm going to convert this
align:start position:0%
sorry I'm going to convert this
Cartesian complex number into an
align:start position:0%
Cartesian complex number into an
align:start position:0%
Cartesian complex number into an
amplitude phase form just because it's
align:start position:0%
amplitude phase form just because it's
align:start position:0%
amplitude phase form just because it's
going to make taking the imaginary part
align:start position:0%
going to make taking the imaginary part
align:start position:0%
going to make taking the imaginary part
of the solution very easy at the end of
align:start position:0%
of the solution very easy at the end of
align:start position:0%
of the solution very easy at the end of
the day
align:start position:0%
align:start position:0%
so the amplitude of this complex number
align:start position:0%
so the amplitude of this complex number
align:start position:0%
so the amplitude of this complex number
is for n sorry 4 minus N squared
align:start position:0%
is for n sorry 4 minus N squared
align:start position:0%
is for n sorry 4 minus N squared
quantity squared
align:start position:0%
quantity squared
align:start position:0%
quantity squared
plus the imaginary part squared so
align:start position:0%
plus the imaginary part squared so
align:start position:0%
plus the imaginary part squared so
that's 4n squared
align:start position:0%
that's 4n squared
align:start position:0%
that's 4n squared
square rooted and
align:start position:0%
square rooted and
align:start position:0%
square rooted and
then we have e
align:start position:0%
then we have e
align:start position:0%
then we have e
upstairs e
align:start position:0%
upstairs e
align:start position:0%
upstairs e
int and downstairs is going to be e I n
align:start position:0%
int and downstairs is going to be e I n
align:start position:0%
int and downstairs is going to be e I n
Phi sorry i phi and
align:start position:0%
Phi sorry i phi and
align:start position:0%
Phi sorry i phi and
i'm going to
align:start position:0%
i'm going to
align:start position:0%
i'm going to
put a subscript n on the phi just
align:start position:0%
put a subscript n on the phi just
align:start position:0%
put a subscript n on the phi just
because for each
align:start position:0%
because for each
align:start position:0%
because for each
for each complex number we're going to
align:start position:0%
for each complex number we're going to
align:start position:0%
for each complex number we're going to
have a different phase Phi and that
align:start position:0%
have a different phase Phi and that
align:start position:0%
have a different phase Phi and that
phase Phi n is
align:start position:0%
phase Phi n is
align:start position:0%
phase Phi n is
going to be the arctangent of
align:start position:0%
going to be the arctangent of
align:start position:0%
going to be the arctangent of
2 n
align:start position:0%
2 n
align:start position:0%
2 n
divided divided by 4 minus n squared
align:start position:0%
align:start position:0%
like that
align:start position:0%
align:start position:0%
okay so now i can combine phi then with
align:start position:0%
okay so now i can combine phi then with
align:start position:0%
okay so now i can combine phi then with
the upstairs term int in the exponent
align:start position:0%
the upstairs term int in the exponent
align:start position:0%
the upstairs term int in the exponent
and when I take the imaginary part I'm
align:start position:0%
and when I take the imaginary part I'm
align:start position:0%
and when I take the imaginary part I'm
only going to be left with sine of n t
align:start position:0%
only going to be left with sine of n t
align:start position:0%
only going to be left with sine of n t
minus Phi n
align:start position:0%
minus Phi n
align:start position:0%
minus Phi n
so X event
align:start position:0%
so X event
align:start position:0%
so X event
yep going is going to be 1 over
align:start position:0%
yep going is going to be 1 over
align:start position:0%
yep going is going to be 1 over
4 minus N squared
align:start position:0%
4 minus N squared
align:start position:0%
4 minus N squared
squared plus 4n squared
align:start position:0%
align:start position:0%
quantity squared square rooted
align:start position:0%
quantity squared square rooted
align:start position:0%
quantity squared square rooted
times sine of n t minus Phi n
align:start position:0%
times sine of n t minus Phi n
align:start position:0%
times sine of n t minus Phi n
where Phi n was given using the
align:start position:0%
where Phi n was given using the
align:start position:0%
where Phi n was given using the
arctangent formula and
align:start position:0%
arctangent formula and
align:start position:0%
arctangent formula and
this gives us a solution which note is
align:start position:0%
this gives us a solution which note is
align:start position:0%
this gives us a solution which note is
periodic with period 2 t.o
align:start position:0%
periodic with period 2 t.o
align:start position:0%
periodic with period 2 t.o
right I i should also note that 5n is
align:start position:0%
right I i should also note that 5n is
align:start position:0%
right I i should also note that 5n is
between
align:start position:0%
align:start position:0%
0 and 2 pi
align:start position:0%
0 and 2 pi
align:start position:0%
0 and 2 pi
sorry 0 it and pie
align:start position:0%
sorry 0 it and pie
align:start position:0%
sorry 0 it and pie
and
align:start position:0%
and
align:start position:0%
and
this gives us this one solution to the
align:start position:0%
this gives us this one solution to the
align:start position:0%
this gives us this one solution to the
differential equation with a forcing of
align:start position:0%
differential equation with a forcing of
align:start position:0%
differential equation with a forcing of
sine NT on the right hand side
align:start position:0%
sine NT on the right hand side
align:start position:0%
sine NT on the right hand side
so now what we want to do is we want to
align:start position:0%
so now what we want to do is we want to
align:start position:0%
so now what we want to do is we want to
sum up many of these solutions using the
align:start position:0%
sum up many of these solutions using the
align:start position:0%
sum up many of these solutions using the
superposition principle so if I go back
align:start position:0%
superposition principle so if I go back
align:start position:0%
superposition principle so if I go back
now
align:start position:0%
now
align:start position:0%
now
i'm going to write the original
align:start position:0%
i'm going to write the original
align:start position:0%
i'm going to write the original
differential equation
align:start position:0%
align:start position:0%
so i'm just going to go back and just
align:start position:0%
so i'm just going to go back and just
align:start position:0%
so i'm just going to go back and just
rewrite this
align:start position:0%
rewrite this
align:start position:0%
rewrite this
and i'm going to write the right-hand
align:start position:0%
and i'm going to write the right-hand
align:start position:0%
and i'm going to write the right-hand
side using its fourier series 1 over n
align:start position:0%
side using its fourier series 1 over n
align:start position:0%
side using its fourier series 1 over n
sine NT
align:start position:0%
sine NT
align:start position:0%
sine NT
where n is odd
align:start position:0%
where n is odd
align:start position:0%
where n is odd
ok and this is the problem we originally
align:start position:0%
ok and this is the problem we originally
align:start position:0%
ok and this is the problem we originally
wanted to solve and essentially what
align:start position:0%
wanted to solve and essentially what
align:start position:0%
wanted to solve and essentially what
we've done is we've solved the problem
align:start position:0%
we've done is we've solved the problem
align:start position:0%
we've done is we've solved the problem
for each individual sine NT so how do we
align:start position:0%
for each individual sine NT so how do we
align:start position:0%
for each individual sine NT so how do we
get the full solution well what we have
align:start position:0%
get the full solution well what we have
align:start position:0%
get the full solution well what we have
to do is we have to multiply the
align:start position:0%
to do is we have to multiply the
align:start position:0%
to do is we have to multiply the
solution for each sign in T by a factor
align:start position:0%
solution for each sign in T by a factor
align:start position:0%
solution for each sign in T by a factor
of 4 divided by PI and 1 over N and then
align:start position:0%
of 4 divided by PI and 1 over N and then
align:start position:0%
of 4 divided by PI and 1 over N and then
we have to add all of these solutions up
align:start position:0%
we have to add all of these solutions up
align:start position:0%
we have to add all of these solutions up
for all odd values of n
align:start position:0%
align:start position:0%
so for example X is just going to be
align:start position:0%
so for example X is just going to be
align:start position:0%
so for example X is just going to be
four over pi
align:start position:0%
four over pi
align:start position:0%
four over pi
sum of n odd
align:start position:0%
sum of n odd
align:start position:0%
sum of n odd
one over n times the solution to every
align:start position:0%
one over n times the solution to every
align:start position:0%
one over n times the solution to every
sine NT
align:start position:0%
sine NT
align:start position:0%
sine NT
which we've already computed and I've
align:start position:0%
which we've already computed and I've
align:start position:0%
which we've already computed and I've
denoted as X sub N and
align:start position:0%
denoted as X sub N and
align:start position:0%
denoted as X sub N and
X sub n is up here
align:start position:0%
X sub n is up here
align:start position:0%
X sub n is up here
so if we want to be explicit about it
align:start position:0%
so if we want to be explicit about it
align:start position:0%
so if we want to be explicit about it
I'll write the whole thing out as
align:start position:0%
I'll write the whole thing out as
align:start position:0%
I'll write the whole thing out as
for over PI and odd
align:start position:0%
align:start position:0%
one over
align:start position:0%
one over
align:start position:0%
one over
four minus N squared quantity squared
align:start position:0%
four minus N squared quantity squared
align:start position:0%
four minus N squared quantity squared
plus 4n squared
align:start position:0%
plus 4n squared
align:start position:0%
plus 4n squared
square rooted
align:start position:0%
square rooted
align:start position:0%
square rooted
we also have a factor of n out front
align:start position:0%
we also have a factor of n out front
align:start position:0%
we also have a factor of n out front
and we have a sine NT
align:start position:0%
and we have a sine NT
align:start position:0%
and we have a sine NT
minus Phi n
align:start position:0%
align:start position:0%
okay so there's the final answer for a
align:start position:0%
okay so there's the final answer for a
align:start position:0%
okay so there's the final answer for a
particular solution in its full glorious
align:start position:0%
particular solution in its full glorious
align:start position:0%
particular solution in its full glorious
detail we can also check that this
align:start position:0%
detail we can also check that this
align:start position:0%
detail we can also check that this
particular solution is periodic note how
align:start position:0%
particular solution is periodic note how
align:start position:0%
particular solution is periodic note how
each sign in T is periodic and each sine
align:start position:0%
each sign in T is periodic and each sine
align:start position:0%
each sign in T is periodic and each sine
NT has period of at least two pi so when
align:start position:0%
NT has period of at least two pi so when
align:start position:0%
NT has period of at least two pi so when
we sum up a whole bunch of functions
align:start position:0%
we sum up a whole bunch of functions
align:start position:0%
we sum up a whole bunch of functions
which all have a period of at least two
align:start position:0%
which all have a period of at least two
align:start position:0%
which all have a period of at least two
pi the the sum is also going to be a
align:start position:0%
pi the the sum is also going to be a
align:start position:0%
pi the the sum is also going to be a
periodic function with at least two pi
align:start position:0%
periodic function with at least two pi
align:start position:0%
periodic function with at least two pi
and
align:start position:0%
and
align:start position:0%
and
and as a result this gives us that the
align:start position:0%
and as a result this gives us that the
align:start position:0%
and as a result this gives us that the
the answer we're looking for also in
align:start position:0%
the answer we're looking for also in
align:start position:0%
the answer we're looking for also in
addition if we want the full general
align:start position:0%
addition if we want the full general
align:start position:0%
addition if we want the full general
solution to the differential equation to
align:start position:0%
solution to the differential equation to
align:start position:0%
solution to the differential equation to
this particular solution we also have to
align:start position:0%
this particular solution we also have to
align:start position:0%
this particular solution we also have to
add the homogeneous piece
align:start position:0%
add the homogeneous piece
align:start position:0%
add the homogeneous piece
so this concludes the problem and I'll
align:start position:0%
so this concludes the problem and I'll
align:start position:0%
so this concludes the problem and I'll
just quickly recap when solving a
align:start position:0%
just quickly recap when solving a
align:start position:0%
just quickly recap when solving a
differential equation with the periodic
align:start position:0%
differential equation with the periodic
align:start position:0%
differential equation with the periodic
forcing function on the right hand side
align:start position:0%
forcing function on the right hand side
align:start position:0%
forcing function on the right hand side
again to iterate the steps you first
align:start position:0%
again to iterate the steps you first
align:start position:0%
again to iterate the steps you first
before you decompose the right hand side
align:start position:0%
before you decompose the right hand side
align:start position:0%
before you decompose the right hand side
into summation of sines and cosines you
align:start position:0%
into summation of sines and cosines you
align:start position:0%
into summation of sines and cosines you
then solve the differential equation for
align:start position:0%
then solve the differential equation for
align:start position:0%
then solve the differential equation for
sine NT cosine NT
align:start position:0%
sine NT cosine NT
align:start position:0%
sine NT cosine NT
individually this gives you a solution
align:start position:0%
individually this gives you a solution
align:start position:0%
individually this gives you a solution
for each term on the right hand side of
align:start position:0%
for each term on the right hand side of
align:start position:0%
for each term on the right hand side of
the differential equation and then at
align:start position:0%
the differential equation and then at
align:start position:0%
the differential equation and then at
the end of the day you superposition to
align:start position:0%
the end of the day you superposition to
align:start position:0%
the end of the day you superposition to
sum up all this all of the solutions and
align:start position:0%
sum up all this all of the solutions and
align:start position:0%
sum up all this all of the solutions and
that gives you one final big solution
align:start position:0%
that gives you one final big solution
align:start position:0%
that gives you one final big solution
okay so i hope you enjoyed this problem
align:start position:0%
okay so i hope you enjoyed this problem
align:start position:0%
okay so i hope you enjoyed this problem
and i'll see you next time