Dynamic
Dynamic
) ( sin ) ( cos 1
) (
) (
0 1
1
) (
0 1
) (
2
0
0
0
0
t t
e
t t e
t H
t x
t t
t t
By subtracting a step function at t , we get an impulse function of width and height a.
x(t)
t
Response of Linear Oscillator to Impulse
a/
0
2
0
= 0.2
0
= 5 (2/
1
)
If we keep the product a constant as we let 0, we get the response to a delta function (see
M&T p. 143-144):
) ( sin
) (
) (
0 1
) (
1
0
t t e
t a
t x
t t
, t > t
0
We can express any forcing function as the sum of impulse functions. If we define F
n
(t) as
a
n
(t) m when t
n
< t < t
n+1
, where t
n+1
t
n
, we can write
t
n
n
t d t t
m
t F
m
t F
m
t F
) (
) ( ) (
lim
) (
0
We know the solution for the nth impulse (it acts over interval t
n+1
t
n
), so we can sum these
to find the solution to the original forcing function:
t
n
n
t t n n
t
n
n
t t e
t a
t x t x ) ( sin
) (
) ( ) (
1
) (
1
0
, t
N
< t < t
N+1
.
Taking the limit as 0 and writing t
n
as t , we get the integral
t d t t e
t a
t x
t t t
) ( sin
) (
) (
1
) (
1
,
and when we say that
'
, 0
), ( sin
1
) , (
1
) (
1
t t e
m t t G
t t
t
t
t < t
(2)
we have
t
t d t t G t F t x ) , ( ) ( ) ( . (3)
Note that while Equation 3 is always true, Equation 2 is specific to this problem: a linear
oscillator initially at rest at equilibrium. The Greens function contains the initial conditions, so
putting any forcing function into Equation 3 will give you an exact solution to the problem.
Though it may be difficult to calculate the integral (as you will see in problem 3-42),
Greens Method always gives you a solution.