[go: up one dir, main page]
Scribd
Upload
202 views3 pages

1 Solving The Damped Harmonic Oscillator Using Green Functions

This document discusses solving the damped harmonic oscillator equation using Green functions. It begins by solving the equation for specific forcing functions like eiωt. It then derives the Green function G(t,t0) which satisfies the differential equation and boundary conditions. The Green function allows writing the general solution as the sum of the homogeneous solution and an integral over G(t,t0) weighted by the forcing function f(t'). Finally, it notes that Green functions are useful for solving inhomogeneous linear partial differential equations and are widely used in physics.

Uploaded by

RincónPrietoAndrés
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
202 views3 pages

1 Solving The Damped Harmonic Oscillator Using Green Functions

This document discusses solving the damped harmonic oscillator equation using Green functions. It begins by solving the equation for specific forcing functions like eiωt. It then derives the Green function G(t,t0) which satisfies the differential equation and boundary conditions. The Green function allows writing the general solution as the sum of the homogeneous solution and an integral over G(t,t0) weighted by the forcing function f(t'). Finally, it notes that Green functions are useful for solving inhomogeneous linear partial differential equations and are widely used in physics.

Uploaded by

RincónPrietoAndrés
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Physics 228, Spring 2001

4/6/01

Solving the damped harmonic oscillator


using Green functions

We wish to solve the equation


y + 2by + !02 y = f (t) .

1.1

The case f (t) / ei!t

First we solve the equation for f (t) =

F (!) i!t
e ,
2

(1)

where F (!) is some number:

F (!) i!t
e
(2)
2
We guess a particular solution of the form yp = cei!t . Plugging in this guess
in order to determine c, we find
y + 2by + !02 y =

cei!t

! 2 + 2ib! + !02 =

which implies a solution for c


c=

F (!) i!t
e ,
2

F (!)/2
.
+ 2ib! + !02

(4)

!2

Therefore the general solution for y with f (t) =


y = yh +

(3)

F (!) i!t
e
2

F (!)/2
ei!t ,
+ 2ib! + !02

!2

is
(5)

and yh is the solution to the homogeneous equation (ie, eq. (1) with f (t) = 0):
yh (t) = C1 er1 t + C2 er2 t ,

(6)

where C1,2 are arbitrary constants and r1,2 are the roots of the quadratic
equation
x2 + 2bx + !02 = 0 .

(7)

In this note I am assuming r1 6= r2 , so that the two exponentials in yh


represent independent solutions.
1

The case f (t) = F (!)ei!t d!


2

1.2

We nowR consider a general forcing function written as a Fourier transform:


. The particular solution now is just the one we found
f (t) = F (!)ei!t d!
2
above, similarly integrated over !. So the general solution is
y = yh +

1.3

!2

F (!)
d!
ei!t
2
+ 2ib! + !0
2

The case f (t) = (t

(8)

t0 )

The Green function for this problem is the function G(t, t0 ) which satisfies
"

d2
d
+ 2b + !02 G(t, t0 ) = (t
2
dt
dt

t0 ) ,

G(t, t0 ) = 0 for t < t0 .

(9)

We can solve for G by writing the Dirac -function (t t0 ) in terms of its


Fourier transform, and then by using the results of the above section:
Z

(t

t0 )e

i!t

=e

i!t0

=)
i!t0

Therefore, from eq. (8), with e


G(t, t0 ) = yh +
= yh +

(t

t0 ) =

i!t0 i!t d!

r1 r2

(10)

replacing F (!), we get

d!
ei!(t t0 )
2
2
! + 2ib! + !0 2
1

er1 (t

t0 )

er2 (t

t0 )

t < t0
t > t0

(11)

The integral I performed above in going from the first line to the second is
easy using complex analysis, hard without it. If we apply the condition that
G(t, t0 ) = 0 for t < t0 , then we can set the constants in yh above to zero.
Therefore our Green function for this problem is:
G(t, t0 ) =

1.4

0
1
r1 r2

r1 (t t0 )

r2 (t t0 )

t < t0
t > t0 .

(12)

Solving the general problem using Green function


techniques

Now we return to the general problem of eq. (1). We can write any function
f (t) as a sum (integral) of delta functions (t t0 ) for dierent values of t0
2

with dierent strengths:


f (t) =

(t

t0 )f (t0 ) dt0

(13)

Thus, since G(t, t0 ) is the particular solution to our dierential equation with
f (t) = (t t0 ), we can construct a particular solution for the general f (t)
by summing up (integrating) G(t, t0 ) for diernt t0 with weight f (t0 ):
yp (t) =

G(t, t0 )f (t0 ) dt0 .

(14)

Note that if f (t) vanishes for t < t1 (ie, the force turns on at time t = t1 ),
then yp (t) similarly vanishes for t < t1 since G(t, t0 ) vanishes for t < t0 and
therefore G(t, t0 )f (t0 ) vanishes for t < t1 . The most general solution then is
y = yh +

G(t, t0 )f (t0 ) dt0 .

(15)

The nice thing about this expression is that G(t, t0 ) and yh are known, so
that one need only perform the integral of your forcing function f (t0 ) times
the known Green function.

1.5

When is this technique useful

Note that we can solve this problem is various ways. However, when trying
to solve linear, second order, inhomogenous partial dierential equations, this
method is the easiest. One can also use the method, along with something
called perturbation theory to solve partial dierential equations which are
almost linear. Green functions are used extensively in many branches of
physics; we will reexamine them when we do partial dierential equations
later this quarter.

You might also like