Solutions to HW 11
(1) (Sec 6.2 Problem 21) y 00 − 2y 0 + 2y = cos(t). y(0) = 1, y 0 (0) = 0
Taking the Laplace’s transform of the differential equation
y − 2y 0 + 2y = cos(t),we have
00
L(y 00 − 2y 0 + 2y) = L(cos(t)) = s2s+1 . Using L(y 00 ) = s2 L(y) − sy(0) − y 0 (0) and
L(y 0 ) = sL(y) − y(0), we have
s2 L(y) − sy(0) − y 0 (0) − 2(sL(y) − y(0)) + 2L(y) = s2s+1 and
(s2 − 2s + 2)L(y) − sy(0) − y 0 (0) + 2y(0) = s2s+1 .
Substituting y(0) = 1 and y 0 (0) = 0, we have
(s2 − 2s + 2)L(y) − s + 2 = s2s+1 and
(s2 − 2s + 2)L(y) = s − 2 + s2s+1 .
s−2 s
So L(y) = s2 −2s+2 + (s2 −2s+2)(s 2 +1) .
Note that s − 2s + 2 = (s − 1)2 + 1. First, we simplify the term by noting that
2
s−2 s−2 −1 3
s2 −2s+2
= (s−1) 2 +1 = (s−1)2 +1 + (s−1)2 +1 .
s
Now we simplify the term (s2 −2s+2)(s 2 +1) by partial fraction. We have
s s a(s−1)+b cs+d
(s2 −2s+2)(s2 +1)
= ((s−1)2 +1)(s2 +1)
= (s−1)2 +1
+ s2 +1
. Multiplying ((s − 1)2 + 1)(s2 + 1),
we have
s = (a(s − 1) + b)(s2 + 1) + (cs + d)((s − 1)2 + 1)
= (as + (−a + b))(s2 + 1) + (cs + d)(s2 − 2s + 2)
= as3 + (−a + b)s2 + as + (−a + b) + cs3 + ds2 − 2cs2 − 2ds + 2cs + 2d
= (a + c)s3 + ((−a + b) + d − 2c)s2 + (a − 2d + 2c)s + (−a + b + 2d). Comparing the
coefficient, we get a + c = 0, −a + b + d − 2c = 0, a − 2d + 2c = 1 and −a + b + 2d = 0.
From a + c = 0, we have c = −a, a + b + d = 0, −a − 2d = 1 and −a + b + 2d = 0.
Multiplying −1 to a + b + d = 0, we get −a − b − d = 0. Adding −a + b + 2d = 0,
we have −2a + d = 0. Using −a − 2d = 1 and −2a + d = 0, we have a = − 51 and
d = − 52 . From a + b + d = 0 and c = −a, we get b = −a − d = 35 and c = 15 . Hence
s (s−1)
(s2 −2s+1)(s2 +1)
= − 51 (s−1) 3 1 1 s 2 1
2 +1 + 5 (s−2)2 +1 + 5 s2 +1 − 5 s2 +1 .
s−2 s
From L(y) = s2 −2s+2 + (s2 −2s+2)(s 2 +1) ,
s−2 s−1 −1
s2 −2s+2
= (s−1)2 +1 + (s−1)2 +1 and
s (s−1)
(s2 −2s+1)(s2 +1)
= − 51 (s−1) 3 1 1 s 2 1
2 +1 + 5 (s−2)2 +1 + 5 s2 +1 − 5 s2 +1 , we get
(s−1)
L(y) = 54 (s−1) 2 1 1 s 2 1
2 +1 − 5 (s−1)2 +1 + 5 s2 +1 − 5 s2 +1 and
y(t) = 45 et cos(t) − 52 et sin(t) + 51 cos(t) − 25 sin(t).
(2) (Sec 6.2 Problem 23) y 00 + 2y 0 + y = 4e−t y(0) = 2, y 0 (0) = −1
Taking the Laplace’s transform of the differential equation y 00 + 2y 0 + y = 4e−t ,
we have L(y 00 + 2y 0 + y) = L(4e−t ). Using L(y 00 ) = s2 L(y) − sy(0) − y 0 (0), L(y 0 ) =
sL(y) − y(0) and L(4e−t ) = 4L(e−t ) = s+14
, we have
2 0 4
s L(y) − sy(0) − y (0) + 2(sL(y) − y(0)) + L(y) = s+1 and
2
(s2 + 2s + 1)L(y) − sy(0) − y 0 (0) − 2y(0) = s+14
.
Substituting y(0) = 2 and y 0 (0) = −1, we have
4 4
(s2 + 2s + 1)L(y) − s · 2 − (−1) − 2 · 2 = s+1 and (s2 + 4s + 4)L(y) = s+1 + 2s + 3.
2 2 2 4
Note that s + 2s + 1 = (s + 1) . So we have (s + 1) L(y) = s+1 + 2s + 3 and
4 2s+3 2s+3
L(y) = (s+1)3 + (s+1)2 . We can simplify (s+1)2 by substituting u = s + 1, s = u − 1
2s+3 2(u−1)+3
and (s+1)2 = u2
= 2u+1
u2
= u2 + u12 = (s+1)
2 1
+ (s+1)2.
4 2 1
Hence we have L(y) = (s+1)3 + (s+1) + (s+1)2 and
y = L−1 ( (s+1)
4 2 1
3 + (s+1) + (s+1)2 ).
Using L(e−t ) = s+11
, L(te−t ) = (s+1)
1 2 −t
2 , and L(t e
2
) = (s+1)3,
t2 e−t
we have L−1 ( s+11
) = e−t , L−1 ( (s+1)
1
2 ) = te
−t
and L−1 ( (s+1)
1
3) = 2
.
Therefore
y = L−1 ( (s+1)
4 2 1
3 + (s+1) + (s+1)2 )
2 e−t
= 4· t 2
+ 2e−t + te−t = 2t2 e−t + 2e−t + te−t .
(3) (Sec 6.3 Problem 8)
0, 0 ≤ t < 1,
f (t) =
t2 − 2t + 2, 1 ≤ t.
We have f (t) = (t2 − 2t + 2)u1 (t). Let g(t − 1) = t2 − 2t + 2 = (t − 1)2 + 1.
Then g(t) = g((t + 1) − 1) = t2 − 1 and f (t) = u1 (t)g(t − 1). Hence L(f (t)) =
L(u1 (t)g(t − 1)) = e−s L(g(t)) = e−s L(t2 + 1) = e−s ( s23 + 1s ).
Sec 6.3(p330) 9
0, 0 ≤ t < π,
f (t) = t − π, π ≤ t < 2π,
0, 2π ≤ t.
We have f (t) = (t−π)uπ,2π (t) = (t−π)(uπ (t)−u2π (t)) = (t−π)uπ (t)−(t−π)(u2π (t).
Let g(t − π) = t − π and h(t − 2π) = t − π Then g(t) = g((t + π) − π) = t,
h(t) = h((t + 2π) − 2π) = t + 2π − π = t + π and f (t) = uπ (t)g(t − π) − u2π (t)g(t − 2π).
Hence L(f (t)) = L(uπ (t)g(t − π) − u2π (t)h(t − 2π)) = e−πs L(g(t)) − e−2πs L(h(t)) =
e−πs L(t) − e−2πs L(t + π) = e−πs s12 − e−2πs ( s12 + π 1s ).
(4) (Sec 6.4 Problem 1) y 00 + y = f (t), y(0) = 0, y 0 (0) = 1
1, 0 ≤ t < π2 ,
f (t) =
0, π2 ≤ t.
1 e−s
We have f (t) = u0 (t) − u1 (t) and L(f (t)) = s
− s
.
3
−s
Since L(y 00 + y) = L(f (t)) = 1s − e s and L(y 00 + y) = (s2 + 1)Y (s) − sy(0) − y 0 (0) =
−s −s
(s2 + 1)Y (s) − 1, we have (s2 + 1)Y (s) − 1 = 1s − e s , (s2 + 1)Y (s) = 1s − e s + 1 and
−s
Y (s) = s(s21+1) − s(se2 +1) + (s21+1) . Using partial fraction, we have s(s21+1) = 1s − (s2s+1)
and g(t) = L−1 ( s(s21+1) ) = 1 − cos(t).
So y(t) = 1−cos(t)+u π2 (t)f (t− π2 )+sin(t) = 1−cos(t)+u π2 (t)(1−cos(t− π2 )+sin(t) =
1 − cos(t) + u π2 (t)(1 − sin(t)) + sin(t).
Note that we have used the fact that cos(t − π2 ) = sin(t).
