Differential Equations
Differential Equations
Differential Equations
Solution #5
1. Solve the following initial value problems with the help of the Laplace transform
Solution:
(a) Applying the Laplace transform to both sides of the differential equation, we
obtain
8 4 10
(s2 Y (s) − 3) + 4Y (s) = 3 − 2 + ,
s s s
so that
1 8 4 10
Y (s) = 2 3+ 3 − 2 +
s +4 s s s
3 2
3s + 10s − 4s + 8
=
s3 (s2 + 4)
2 1 2 s 2
= 3 − 2 + −2 2 +2 2 .
s s s s +4 s +2
Applying the inverse Laplace transform yields
(b) Applying the Laplace transform to both sides of the differential equation yields
2 2 s2 + 1
s2 W (s) + s − 1 + W (s) = + = 2
s3 s s3
and
s2 + 1
(s2 + 1)W (s) = −s + 1 + 2
s3
and finally
s 1 2
W (s) = − + 2 + 3.
s2 +1 s +1 s
Hence, we have
w(t) = − cos t + sin t + t2 .
1
(c) Define z(t) := y(t + π). Then the initial value problem becomes
z 00 + z = t + π, z(0) = 0, z 0 (0) = 0.
so that
i.e.,
(s3 − s2 + s − 1)Y (s) = s2 + 3.
2
Solution: Apply the Laplace transform to both sides of the differential equation:
2
L{ty 00 }(s) − L{ty 0 }(s) + L{y}(s) = .
s
Note that
so that
d d
L{ty 0 }(s) = − L{y 0 }(s) = − (sY (s) − 2) = −sY 0 (s) − Y (s)
ds ds
and
d d
L{ty 00 }(s) = − L{y 00 }(s) = − (s2 Y (s) − 2s + 1) = −s2 Y 0 (s) − 2sY (s) + 2.
ds ds
Altogether, we obtain
2
(−s2 Y 0 (s) − 2sY (s) + 2) − (−sY 0 (s) − Y (s)) + Y (s) =
s
or, equivalently,
2(1 − s)
s(1 − s)Y 0 (s) + 2(1 − s)Y (s) = .
s
Division by s(1 − s) yields the fird order linear differential equation
2 2
Y 0 (s) + Y (s) = 2 .
s s
Its integrating factor is
Z
2
µ(s) = exp ds = e2 ln |s| = s2 ,
s
so that Z Z
1 2 1 2 C
Y (s) = µ(s) 2
ds = 2 2 ds = +
µ(s) s s s s2
for some constant C. It follows that
y(t) = 2 + Ct.
4. The current I(t) in an LC series circuit is governed by the initial value problem
where (
3 sin t, 0 ≤ t ≤ 2π,
g(t) :=
0, 2π < t
3
Determine the current as a function of time.
Solution: Since
L{I 00 }(s) = sL{I}(s) − s − 3,
we obtain
so that
e−2πs 3(1 − e−2πs )
1
L{g}(s) = 3 − = .
s2 + 1 s2 + 1 s2 + 1
Solving for L{I}(s) yields
s 3 3(1 − e−2πs )
L{I}(s) = + + .
s2 + 4 s2 + 4 (s2 + 1)(s2 + 4)
Since
3 1 1
= 2 − 2 ,
(s2 2
+ 1)(s + 4) s +1 s +4
we obtain
s 3 1 1
L{I}(s) = 2 + 2 + (1 − e−2πs ) 2
− 2
s +4 s +4 s +1 s +4
and thus
3 1 1
I(t) = cos 2t + sin 2t + sin t − sin 2t − sin(t − 2π) − sin 2(t − 2π) u(t − 2π)
2 2 2
1
= sin t + sin 2t + cos 2t + sin 2t − sin t u(t − 2π).
2
Determine L{f }.
Solution: The windowed version fT of f is
t,
0 < t < 1,
fT (t) := 1 − t, 1 < t < 2,
0, 2 < t.
4
Its Laplace transform FT is computed as (integration by parts)
Z ∞
FT (s) = e−st fT (t) dt
0
Z 1 Z 2
−st
= e t dt + e−st (1 − t)
0 1
1 − 2e−s − se−s + e−2s + se−2s
=
s2
It follows that
FT (s) 1 − 2e−s − se−s + e−2s + se−2s
L{f }(s) = = .
1 − e−2s s2 (1 − e−2s )
6. Use convolution to determine a formula for the solution of the initial value problem
so that
(s2 + 9)Y (s) = s + G(s)
and thus
s 1 3
Y (s) = + G(s).
s2
+ 9 3 s2 + 1
Applying the inverse Laplace transform, we obtain
−1 s 1 −1 3
y(t) = L (t) + L G(s)
s2 + 9 3 s2 + 1
1
= cos 3t + (sin 3t) ∗ g(t)
3
Z t
= cos 3t + sin(3(t − τ ))g(τ ) dτ.
0
7. Use convolution to find the inverse Laplace transforms of the following functions:
1 1 s+1
(a) ; (b) ; (c) .
(s + 1)(s + 2) s3 (s2 + 1) (s2 + 1)2
Solution:
5
n o
(a) Since L−1 1
s−a (t) = eat , we have
−1 1 −1 1 1
L (t) = L (t)
(s + 1)(s + 2) s+1s+2
−1 1 −1 1
=L (t) ∗ L (t)
s+1 s+2
= e−t ∗ e−2t
Z t
= e−(t−τ ) e−2τ dτ
0
Z t
−t
=e e−τ dτ
0
= e (1 − e−t )
−t
= e−t − e−2t .
(b) We have
−1 1 −1 1 1
L (t) = L (t)
s3 (s2 + 1) s3 s2 + 1
−1 1 −1 1
=L (t) ∗ L (t)
s3 s2 + 1
t2
= ∗ sin t
2
1 t
Z
= (t − τ )2 sin τ dτ
2 0
Z t
1 2
t
= −(τ − t) cos τ 0 + 2 (τ − t) cos τ dτ
2 0
Z t
1 2 t
= t + 2(τ − t) sin τ |0 − 2 sin τ dτ
2 0
t2
= + cos t − 1.
2
6
For the first summand, we have
−1 s −1 s −1 1
L (t) = L (t) ∗ L (t)
(s2 + 1)2 s2 + 1 s2 + 1
= cos t ∗ sin t
Z t
= cos(t − τ ) sin τ τ
0
1 t
Z
= sin t + sin(t − 2τ ) dτ
2 0
!
1 cos(t − 2τ ) t
= τ sin t +
2 2
0
t sin t
= .
2
Altogether, we obtain
−1 s+1 t sin t sin t − t cos t t sin t + sin t − t cos t
L 2 2
(t) = + = .
(s + 1) 2 2 2
so that
t 3 3e2t
y(t) = − + 2e−t + .
2 4 4
9. A linear system is governed by the initial value problem
Find the transfer function H(s) of the system, the impulse response function h(t),
and give a formulate for the solution to the initial value problem.
7
Solution: The transfer function is
1 1
H(s) = = .
s2 + 2s − 15 (s − 3)(s55)
Consequently, the impulse response function is computed as
h(t) = L−1 {H}(t)
−1 1 −1 1
=L (t) ∗ L (t)
s−3 s+5
= e3t ∗ e−5t
Z t
= e3(t−τ ) e−5τ dτ
0
−8τ t
!
e
= e3t −
8 0
e3t − e−5t
= .
8
To solve the initial value problem, we need the solution yk (t) for the homogeneous
problem
y 00 + 2y 0 − 15y = 0, y(0) = 0, y 0 (0) = 8.
Applying the Laplace transform yields
so that
8
Yk (s) = .
s2
+ 2s − 15
Since Yk (s) = 8H(s), it follows that yk (t) = 8h(t) = e3t − e−5t . All in all, we obtain
1 t 3(t−τ )
Z
y(t) = (h ∗ g)(t) + yk (t) = (e − e−5(t−τ ) )g(τ ) dτ + e3t − e−5t .
8 0
10. Use convolution to show that
Z t
−1 F (s)
L (t) = f (τ ) dτ
s 0
8
11. Solve the symbolic initial value problem
Solution: Applying the Laplace transform to both sides of the equation, we obtain
12. A mass attached to a spring is released from rest 1 meter below the equilibrum
position for the mass-spring system and begins to vibrate. After π2 seconds, the mass
is struck by a hammer exerting an impulse on the mass. The system is governed by
the symbolic initial value problem
d2 x π dx
2
+ 9x = −3δ t − , x(0) = 1, (0) = 0
dt 2 dt
where x(t) denotes the displacement from equilibrum at time t. What happens to
the mass after it has been struck?
Solution: Applying the Laplace transform, we obtain
π
s2 X(s) − s + 9X(s) = −3e− 2 s ,
so that
s π 3
X(s) = − e− 2 s 2 .
s2 +9 s +9
It follows that
π π π
x(t) = cos 3t − sin 3 t − u t− = cos 3t − cos 3t u t − ,
2 2 2
i.e., (
cos t, 0 < t < π2 ,
x(t) = π
0, 2 <t
The blow of the hammer thus stops the mass in its equilibrum position.
9
13. Use Laplace transforms to solve the initial value problem
Solution: Applying the Laplace transform to the two differential equations yields
e−3s
sX(s) + Y (s) = X(s) and 2sX(s) + s2 Y (s) − s + 1 =
s
i.e.,
e−3s
(s − 1)X(s) + Y (s) = 0 and 2sX(s) + s2 Y (s) = s − 1 + .
s
Solving for X(s) yields
1 − s − e−3s
X(s) =
s(s2 − s − 2)
1−s 1
= − e−3s
s(s + 1)(s − 2) s(s + 1)(s − 2)
11 2 1 1 1 −3s 1 1 1 1 1 1 1 1
=− + − −e − − +
2s 3s+1 6s−2 4 s 2 s2 3 s + 1 12 s − 2
and thus
1 2e−t e2t 1 t − 3 e3−t e2t−6
x(t) = − + − − − − + u(t − 3).
2 3 6 4 2 3 12
10