MATH 2003 - DIFFERENTIAL EQUATIONS

CHAPTER 1: TUTORIAL (Questions marked with an asterisk are optional.)

1. Verify that
i) y = sin 2x ii) y = cos 2x
iii) y = A sin 2x + B cos 2x iv) y = R cos(2x − α)
d2 y
are all solutions of the differential equation + 4y = 0, where A, B, R, α are
dx2
constants.

Solve the following differential equations. Whenever a condition is given find the
particular solution.
dy
2. = 2xy .
dx
dy
3. xy = 1 − x2 .
dx
√
4. (1 + 3x2 ) dy = (1 − 3y 2 ) dx, given that y = 3 when x = 1 .
dy
5. = (3x + 2)−2 .
dx
dy
6. y(x + 1) = x2 + 2 .
dx
dy
7. = y2, given that y(0) = 1 .
dx
dy 2y sin x
8. + = .
dx x x
9. f ′ (x) − f (x) sec x = ex cos x, given that f (0) = 1 .

10. y ′ = x(x2 + 2y) .

dy y
11. − = x sec2 x .
dx x
12. y ′ + y tan x = cos x, given that y(0) = 2 .

13. x2 dy − sin 3x dx + 2xy dx = 0 .

1
 dy
14. 2x + y 3 + (3xy 2 + ye2y ) = 0.
dx
dy
15. y cos x + 2xey + x sin x + (sin x + x2 ey + 2y) = 0.
dx
16. (2x + 3y + 1) dx + (3x + 4y − 1) dy = 0 .

17. x(x + y) dy − y 2 dx = 0 .

18. x dy − y dx + xe−y/x dx = 0 ,

19. (x2 − y 2 ) dx + (x2 + 2xy) dy = 0 .

dy x2 y 2
20. dx
= 1+x
.

21. 2y(x + 1) dy = x dx .
dy
22. dx
= sin x(cos 2y − cos2 y) .
1
! 2 3 " dx
23. x y − 1+9x 2 dy
+ x3 y 2 = 0 .

24. (y 2 cos x−3x2 y −2x) dx+(2y sin x−x3 +ln y) dy = 0, given that y(0) = e .

25. (y 2 + xy) dx − x2 dy = 0 .
√
26. −y dx + (x + xy) dy = 0 .
# %
dy
$ x dx
27. x + y 2 − xy dx = y, given that y(1/2) = 1 . [Hint: Put v = y
and use dy
].

dy
28. x dx − 4y = x6 ex .
dx
29. (x+4y 2) dy +2y dx = 0 . [Hint: Use the form for a linear d.e.: dy
+xP (y) =
Q(y)].

30. cos2 x sin x dy + (y cos3 x − 1) dx = 0 .

dy y
31. dx
= y−x
, given that y(5) = 2 .
dy
32. x2 dx + y 2 = xy .
dy
33. (x + 1) dx + y = ln x, given that y(1) = 10 .
d2 y
! dy "2 dy
34. dx 2 = 2x dx
, [Hint: put dx = u] .

2
 dy d2 y du du dy du
35. y ′′ + 2y(y ′)3 = 0 , [Hint: Put u = dx
, so that dx2
= dx
= dy dx
= dy
u].

*36. There is another type of first order d.e. called a Bernoulli equation. It is of
the form
dy
+ y P (t) = y n Q(t) .
dt
If n = 0, it is linear, and if n = 1 it is separable.
The method to solve such d.e. when n ̸= 0, and n ̸= 1 is as follows:
Multiply the d.e. by y −n (1 − n). The d.e. then becomes:

d 1−n
(y ) + (1 − n)y 1−n P (t) = (1 − n)Q(t) .
dt
(Check this for yourself.)
Use the substitution z = y 1−n to solve the linear d.e.:
dz
+ (1 − n)P (t)z = (1 − n)Q(t) .
dt
Hence solve for y.
dy 2
Use the above method to solve + yt = te−t y −3 .
dt

ANSWERS TO CHAPTER 1 TUTORIAL

2
2. y = Aex .
3. y 2 =& ln x2 −& x2 + C .
1 & 1+√3 y & √
4. ln & 1−√3 y & = arctan 3 x + C.
2
1 && 1+√3 y && √ √ π
The particular solution is ln & 1−√3 y & = arctan 3 x + ln 2 − .
2 3
−1
5. y = +C.
3(3x + 2)
6. y 2 = x2 − 2x + 6 ln |x + 1| + C .
7. y(x + C) + 1 = 0 . The particular solution is y(x − 1) + 1 = 0 .
8. x2 y = sin x − x cos x + C .
cos x 1
9. f (x) = ex + ex (cos x − sin x) + C .
1 + sin x 2
cos x 1 1
The particular solution is f (x) = ex + ex (cos x − sin x) − .
1 + sin x 2 2
x2 + 1 2
10. y = − + Cex .
2
11. y = x tan x + xC .
12. y sec x = x + C . The particular solution is y sec x = x + 2 .

3
 cos 3x
13. x2 y + =C.
3
ye2y e2y
14. x2 + xy 3 + − =C.
2 4
15. y sin x + x2 ey + y 2 − x cos x + sin x = C .
16. x2 + 2y 2 + x + 3xy − y = C .
17. y = C e −y/x .
18. e y/x = − ln |x| + C .
19. y 2 + xy + x2 = C x .
x2
20. 2
− x + ln |1 + x| = C − y1 .
21. y 2 = x − ln |x + 1| + C .
22. − cot y = cos x + C .
23. x3 y 3 − arctan 3x = C .
24. y 2 sin x − x3 y − x2 + y ln y − y = 0 .
25. x + y ln |x|
'= Cy .
26. ln |y| = 2 xy + C .
' √
27. ln |y| = −2 1 − xy + 2 .
28. y = x5 ex − x4 ex + Cx4 .
1
29. x = − 54 y 2 + Cy − 2 .
30. y = sec x + Ccosec x .
31. x = y2 + y8 .
x
32. e y = Cx .
33. (x + 1)y = x ln x − x + 21 .
34. y + C2 = − C11 arctan Cx1 .
y3
35. 3
− C1 y = x + C2 .
2 2
36. y 4 = 2e−t + Ce−2t .

4
CHAPTER 2: TUTORIAL
1. Show that if y1 = µ−1 µq(t) dt, where µ = e p(t) dt , then y1 is a solution of the
(

first order linear differential equation y ′ + p(t)y = q(t).

(
(Hint: write µy1 = µq(t) dt, then differentiate both sides, and use the fact that
p(t) = µ′ /µ.)

2. Solve the first order linear differential equations below, and show that the general
solution is in the sum of a particular solution of the given equation and a general
solution of the homogeneous equation.

2 dy y 1
(a) y ′ + 2ty = e−t (b) − 2 = 2 .
dt t +1 t +1
3. The following list gives the outputs P (y) obtained when different operators P
act on the general function y = y(t).
Simplify P (y1 + y2 ) and P (y1) + P (y2 ), and see if they are equal.
Similarly, compare P (cy) (where c is constant) and cP (y).
Hence determine which operators are linear.

(a) P (y) = 2y , (b) P (y) = y 2 , (c) P (y) = yy ′ , (d) P (y) = ty ′ +y.

4. Use the rules for differentiation and integration to show that the differentiation
(b
operator (y → → y ′ ) and the definite integral operator (y → → a y dt) are
linear operators.
5. Find the null spaces of the linear operators in Question 3.
(Hint: put the given output equal to 0, and solve for y.)

6. Find (D 2 + 3D + 2)(cy) and (D 2 + 3D + 2)(y1 + y2 ), where c is constant. Hence

show that D 2 + 3D + 2 is a linear operator.
*Show that a general D-operator D n + an−1 D n−1 + · · · + a0 is linear.

7. Verify that ( dtd + 2) ( dtd + 1)y and ( dtd + 1) ( dtd + 2)y are both equal to (D 2 +
! " ! "

3D + 2)y.

8. Show that D(D + t)y ̸= (D + t)Dy. (Hint: simplify each side separately, and
don’t forget the product rule for differentiation, when it is required.)

9. Use Question 1 to verify that y1 = eαt e−αt f (t) dt is a particular solution of the
(

5
equation (D − α)y = f (t), and that y0 = Aeαt is a general function in the null space.
(Hint: put p(t) = −α, and replace q(t) by f (t) and by zero.)

10. Use repeated first order methods to solve the D-operator equations:
d2 y dy
(a) (D 2 + 3D + 2)y = 0 (b) 2 − 4 + 4y = 0 (c) ÿ + 3ẏ + 2y = 4t.
dt dt
11. Write down complementary functions for the D-operators:
d3 d2 d
(a) (D + 2)(D − 1)2 , (b) D 2 (D − 2)3 , (c) 3 + 2 − − 1 , (d) (D 2 − 1)2 .
dt dt dt
12. Find complementary functions and particular integrals for the equations from
Question 10, and compare the solutions with those obtained previously.

13. Find the complete general solutions of the D-operator equations below.
(Hint: complementary functions were found in Question 11.)
(a) (D + 2)(D − 1)2 y = e−t (b) D 2 (D − 2)3 y = et
) 2 *2
d
(c) (D 3 + D 2 − D − 1) y = t (d) − 1 y = t3 .
dt2
*14 Use integration by parts several times (preferably in one line) to show that
, -
1 f ′ (t) f ′′ (t)
+
αt −αt
e e f (t) dt = − f (t) + + +··· .
α α α2
We know from Question 9 that the left hand side is a particular solution of (D−α)y =
1
f (t). Show that the right hand side can be obtained by expanding D−α as a bino-
mial series and applying it to f (t). (This justifies the series expansion of an inverse
D-operator, provided that the right hand side converges.)

15. Find complete general solutions of the equations below.

(a) (D + 2)(D − 1)2 y = 6et (b) D 2 (D − 2)3 y = 4te2t
3 2
(c) ddt3y + ddt2y − dy
dt
− y = 8 cosh t (d) (D 2 − 1)2 y = 4tet .
16. Find real-valued solutions for the following homogeneous D-operator equations.
d4 y
(a) (D 2 + 4)y = 0 (b) + 4y = 0 (c) (D 2 + 4D + 5)y = 0
dt4
(d) (D 3 + D 2 + D + 1)y = 0 (e) ÿ + 2ẏ + 2y = 0.

17. Find particular integrals for the following D-operator equations. Hence write
down the complete general solutions. (Complementary functions were found in
Question 16.)

6
 (a) ÿ + 4y = t2 + e−t (b) (D 4 + 4)y = sin t
2
(c) ddt2y + 4 dy
dt
+ 5y = 8t sin t (d) (D 3 + D 2 + D + 1)y = te−t
2
(e) (D + 2D + 2)y = 5 cos t.

18. Find real general solutions of the simultaneous differential equations below by
the following method: differentiate the first equation to get ẍ in terms of ẋ and ẏ,
and substitute for ẏ so that ẍ is expressed in terms of ẋ, x, and y. Then eliminate y
from the expressions for ẍ and ẋ, and solve the resulting D-operator equation for x.
Finally, find y by using the first equation to express y in terms of x and ẋ.
(Dots indicate derivatives with respect to t.)
ẋ = y ẋ = −x + y
(a) (b)
ẏ = −x ẏ = −x − y.
19. Solve the following homogeneous differential equations under the given initial
conditions:

d2 y(t)
&
dy(t) dy(t) &&
(a) + 5 + 6y(t) = 0, y(0) = 1, = 0.
dt2 dt dt &t=0
d2 y(t)
&
dy(t) dy(t) &&
(b) + 2 + y(t) = 0, y(0) = 1, = 0.
dt2 dt dt & t=0
&
(c) (D 2 + 1)y = 0, y(0) = 1, D(t)&t=0
= 1.

(d) ẍ + 4ẋ + 13x = 0 such that x = 2 and ẋ = 1 when t = 0.

d4 y dy d2 y d3 y
(e) 4 − y = 0 such that y = 1 and = 2 = 3 = 0 when t = 0.
dt dt dt dt

20. Solve the following differential equations, using the given initial conditions to
find the constants.

d2 y(t)
&
dy(t) dy(t) &&
(a) 2
+3 + 2y(t) = e−t , y(0) = = 0.
dt dt dt &t=0
&
(b) (D 2 + 3D + 2)y = e−t , y(0) = D(t)&t=0 = 1.

d2 y(t)
&
dy(t) dy(t) &&
(c) + 2 + y(t) = te−t , y(0) = 1, = 0.
dt2 dt dt &t=0

7
(d) ẍ(t) − x(t) = e−t + sin t, x(0) = ẋ(0) = 0.

21. If P (D) = (D + 1)2 and Q(D) = D − 1, find the full general solution of
P (D)y = Q(D)te−t . (Hint: First simplify the right hand side.)

ANSWERS TO CHAPTER 2 TUTORIAL

(
1. µy1 = µq(t) dt, so µy1′ + µ′ y1 = µq(t). Now substitute for µ′ and cancel µ.

2 2
2. (a) y = y1 + y0 , where y1 = te−t and y0 = ce−t . Check that y0′ + 2ty0 = 0.

y0
(b) y = y1 + y0 , where y1 = −1 and y0 = cearctan t . Check that y0′ − t2 +1
= 0.

3. (a) P (cy) = 2cy = c(P y), P (y1 + y2 ) = 2(y1 + y2 ) = P y1 + y2 , linear;

(b) P (cy) = c2 y 2 ̸= cy 2 = c(P y);
P (y1 + y2 ) = (y1 + y2 )2 ̸= y12 + y22 = P y1 + P y2 , non-linear ;
(c) P (cy) = c2 yy ′ ̸= cyy ′ = c(P y),
P (y1 + y2 ) = (y1 + y2 )(y1′ + y2′ ) ̸= y1 y1′ + y2 y2′ = P y1 + P y2, non-linear ;
(d) P (cy) = tcy ′ + cy = c(P y),
P (y1 + y2) = t(y1′ + y2′ ) + (y1 + y2 ) = (ty1′ + y1) + (ty2′ + y2) = P y1 + P y2, linear.

c
5. (a) y = 0, (d) y = .
t

7. ( dtd + 2) ( dtd + 1)y = ( dtd + 2)( dy

! "
dt
+ y)
d2 y
= dt ( dt + y) + 2( dt + y) = ( dt2 + dy
d dy dy
dt
) + (2 dy
dt
+ 2y)
2
d y dy 2
= dt2 + 3 dt + 2y = (D + 3D + 2)y.

8. D(D + t)y = D(y ′ + ty) = y ′′ + (ty ′ + y) (product rule), while

(D + t)Dy = (D + t)y ′ = y ′′ + ty ′ .
This shows that if the coefficients are not constants, then the operators cannot
simply be multiplied like polynomials.

10. (a) y = Ae−2t + Be−t , (b) y = (At + B)e2t , (c) y = 2t − 3 + Ae−2t + Be−t .

8
11. (a) Ae−2t + (B1 + B2 t)et ,
(b) A1 + A2 t + (B1 + B2 t + B3 t2 )e2t ,
(c) Aet + (B1 + B2 t)e−t ,
(d) (A1 + A2 t)et + (B1 + B2 t)e−t .

13. (a) 14 e−t + Ae−2t + (B1 + B2 t)et , (b) −et + A1 + A2 t + (B1 + B2 t + B3 t2 )e2t ,
(c) −t + 1 + Aet + (B1 + B2 t)e−t , (d) t3 + 12t + (A1 + A2 t)et + (B1 + B2 t)e−t .

e−αt e−αt ′ e−αt ′′

(
*14 e−αt f (t) dt = −α
f (t) − (−α)2
f (t) + (−α)3
f (t) − ···.

15. (a) t2 et + Ae−2t + (B1 + B2 t)et ,

1 4
(b) ( 24 t − 16 t3 )e2t + A1 + A2 t + (B1 + B2 t + B3 t2 )e2t ,
(c) tet − t2 e−t + Aet + (B1 + B2 t)e−t ,
(d) ( 61 t3 − 12 t2 )et + (A1 + A2 t)et + (B1 + B2 t)e−t .

16. (a) y0 = P cos 2t + Q sin 2t or y0 = R cos(2t − γ),

(b) y0 = et (P cos t + Q sin t) + e−t (R cos t + S sin t),
(c) y0 = e−2t (P cos t + Q sin t),
(d) y0 = Ae−t + P cos t + Q sin t,
(e) y0 = e−t (P cos t + Q sin t).

17. (a) y = 14 (t2 − 12 ) + 15 e−t + P cos 2t + Q sin 2t,

(b) y = 15 sin t + et (P cos t + Q sin t) + e−t (R cos t + S sin t),
(c) y = cos t(−t + 1) + sin t(t − 12 ) + e−2t (P cos t + Q sin t),
(d) y = ( 14 t2 + 12 t)e−t + Ae−t + P cos t + Q sin t,
(e) y = cos t + 2 sin t + e−t (P cos t + Q sin t).

18. (a) x = P cos t + Q sin t, y = Q cos t − P sin t,

(b) x = e−t (P cos t + Q sin t), y = e−t (Q cos t − P sin t),

9
19. (a) y(t) = 3e−2t − 2e−3t ,
(b) y(t) = e−t + te−t ,
(c) y(t) = sin t + cos t,
(d) x = e−2t (2 cos 3t + 35 sin 3t),
(e) y(t) = 12 (cosh t + cos t).

20. (a) y(t) = e−2t + (t − 1)e−t ,

(b) y(t) = −e−2t + (t + 2)e−t ,
(c) y(t) = e−t + (te−t + 16 t3 e−t ,
(d) x(t) = 12 (et − e−t − te−t − sin t) .

21. Q(D)te−t = (1 − 2t)e−t . Solution is y = e−t (A + Bt + 21 t2 − 13 t3 ).

10
CHAPTER 3: TUTORIAL (Questions marked with an asterisk are for enrich-
ment.)
(b
1. Write down the definition of a f (t) dt if the integral is improper at t = a only.

(∞
*2. If σ > a, prove that T0
Ke−(σ−a)t dt converges.

(∞
*3. (a) Show that 1 x−p dx converges for p > 1. Evaluate the integral in terms
of p, and indicate the region whose area is represented by the integral on a sketch
of the graph of y = x−p for p > 1.

(1
(b) Similarly, show that 0 x−q dx converges for 0 < q < 1. Evaluate the integral in
terms of q and indicate the corresponding region on the sketch.

(c) Substitute q = p1 and show that the value of the integral in (b) is 1 plus the value
of the integral in (a). Use the sketch to explain why. (Hint: the graphs y = x−p
for x ≥ 1 and y = x−1/p for 0 < x ≤ 1 are reflections of each other in the line y = x.)

4. Use formulae from the notes to find the Laplace transforms of the following
functions. (Where necessary, express the given functions in terms of others whose
transforms are known, and use linearity.)
(a) e2t (b) e−t (c) cosh 2t (d) cos t
(e) sin(t + π3 ) (f) e2t sin t (g) e−t cos 12 t (h) t2
(i) (t + 1)3 (j) te−t (k) t2 sinh 2t (l) tet sin t.

5. Find the Laplace transforms of cosh bt and sinh bt, and compare them with those
of cos bt and sin bt.

6. (a) Take Laplace transforms of the terms in the Maclaurin series for et , using
sigma notation and the formula for L(tn ). Sum the resulting (geometric) series in s,
and hence confirm the formula for L(et ).
(b) Repeat (a) with sin t in place of et .
(c)* Similarly, find L sint t .
! "

7. Use the definition of Laplace transform to find the Laplace transforms of the
following functions by direct integration:

11
 ⎧ ⎧
⎨t if 0 ≤ t ≤ T ⎨sin t if 0 ≤ t ≤ π
(a) f (t) = (b) g(t) =
⎩0 if t > T ⎩0 if t > π.
! "
8. Use the definition of Laplace transform to show that if L f (t) = F (s) and a > 0,
then L f (at) = a1 F ( as ). (Hint: substitute u = at in the integral.)
! "

Verify this formula when f (t) = et , using transform tables.

! "
9. Use the definition of Laplace transform to show that if F (s) = L f (t) , then
(∞
lims→0 F (s) = 0 f (t) dt, assuming that the right hand side exists.
Verify this formula by evaluating both sides when f (t) = e−t .
If f (t) = cos t, show that the limit on the left hand side exists, but the integral on
(T
the right hand side does not converge. (Hint: evaluate 0 cos t dt; what happens as
T → ∞?)

10. By investigating limits as s → ∞, s ∈ R, determine which of the following

definitely do not have an inverse transform.
2
(a) s + 1 (b) ss−1
2 +1 (c) ss2 −1
+1
(d) ln s−1
s+1
.

11. Invert the following transforms, using partial fractions and table of transforms
1 1 8 2
(a) 2 (b) 2 (c) (d)
s −1 s − 2s s(s2 + 4) s(s2 + 2s + 2)
1 1 4 1
(e) 2 2
(f) 2 (g) 2 2 (h) 3 .
(s − 1) !s ′(s −
" 1) ! " s (s + 4) s (s + 1)
12. Verify the formula L f (t) = sL f (t) − f (0) when f (t) = cos bt.

13. If f (t) = tn , where n is a natural number, show that

f (0) = f ′ (0) = · · · = f (n−1) (0) = 0

and that
f (n) (t) = n!.
r
[(Hint: dtd r (tn ) = n(n−1) . . . (n−r+1)tn−r .]
Substitute these in the formula for L f (n) (t) and hence give a second proof of the
! "

formula for L tn .
! "

14. Use Laplace transforms to find general solutions of the following D-operator
equations.

12
 (a) (D + 1)y = t (b) (D + 1)y = e−t (c) (D 2 − 1)y = 2t.
Determine the particular solution of (c) satisfying y(0) = 1 and y ′(0) = 0, and the
particular solution satisfying y(0) = y(1) = 0.

15. A particle, initially at rest at the origin, undergoes forced damped oscillations
with equation
y ′′ + 4y ′ + 5y = 5 .
Find its displacement at time t. The particle at rest at the origin means that
y(0) = y ′ (0) = 0.

16. Use Laplace transforms to solve the following pair of simultaneous differential
equations, given y(0) = z(0) = 0. (Hint: after transforming, use matrix methods to
solve the resulting simultaneous algebraic equations.)

y +2ẏ +3z +ż = t + 5

3y +ẏ −z −2ż = 3t.
! "
17. By writing sF (s) = L f ′ (t) + f (0) (assuming f ′ (t) has a Laplace transform),
and applying Question 9 and other results to f ′ (t), prove that

lim sF (s) = lim f (t) and lim sF (s) = lim f (t),

s→∞ t→0 s→0 t→∞

when the limits on both sides exist.

18. By putting f (t) = 1 in the formula for L tn f (t) , obtain a third proof of the
! "

formula for L tn .
! "

19. Show that L sint t = π2 − arctan s.

! "

Hint: Let f (t) = sint t , multiply through by t, apply the theorem, and then integrate
with respect to s. Evaluate the constant of integration by considering lims→∞ F (s).
Use elementary trigonometry to show that the right hand side is equal to arctan 1s
! "
(∞
if s > 0. By letting s → 0, show using Question 9 that 0 sint t dt = π2 .

20. Find L−1 ln s+a . [Hint: let F (s) = ln s+a d

1 ! "2 ! "
s+b s+b
, and find − ds F (s); then invert,
using tables, to find tf (t).]

13
 1 2
Similarly, find L−1 ln(1 + s−2 ) .

21. If f (t) = ln t, show that limt→0 tf (t) = 0 (L’Hôpital) and that dtd (tf (t)) = 1 +
d
f (t). By taking Laplace transforms of this equation, show that s ds F (s)+F (s) = − 1s .
Solve the differential equation and deduce that L(ln t) = − 1s (γ + ln s), where γ is a
constant, called Euler’s constant.

22. Prove that L f (t)

! " (∞
t
= s F (p) dp by showing separately that each side tends to
zero as s → ∞ and has derivative −F (s). Use this formula to check the answer of
Question 19.

23. Use Laplace transforms to solve the differential equations:

(i) ty ′′ + y ′ = 1 + t, given y = 2 when t = 0.
(ii) ty ′′ − (t + 2)y ′ + 3y = t − 1, given y(0) = 0 and y(1) = 23 .

24. If y(t) is a solution of the differential equation ty ′′ + y ′ + ty = 0, and if y and y ′

are defined at t = 0, show that Y (s) = A(s2 + 1)−1/2 , where A is an arbitrary con-
stant, and that y(0) = A. (The solution with A = 1 is called Bessel’s function J0 (t).)

25. Use Questions 9 and 22 to show that 0 f (t)

(∞ (∞
t
dt = 0 F (s) ds, provided both
(∞
sides exist. Hence give another proof that 0 sint t dt = π2 . (See Question 19.)
26. Solve for f (t), the integral equation
+ t
2
f (t) = 2t + sin 4x f (t − x) dx .
0

27. Prove that (f ∗ g)(t) = (g ∗ f )(t) by substituting v = t − u in the integral

(t
(f ∗ g)(t) = 0 f (u)g(t − u) du.

!( t " 1
! "
28. Using tables, verify the formula L 0
f (u) du = s
L f (t) with the functions
f (t) = tn and f (t) = sin bt.

29. Use tables to verify the formula L(f ∗g) = L(f )L(g) with f (t) = t and g(t) = et .

14
ANSWERS TO CHAPTER 3 TUTORIAL
(b (b
1. a f (t) dt = limh→a+ h f (t) dt.

(T
2. T0 Ke−(σ−a)t dt = K e , which tends to K e σ−a 0 as T → ∞
−(σ−a)T0 −e−(σ−a)T −(σ−a)T
σ−a
(negative exponential).

1 1
3. (a) p−1 {1 − h1−p } → p−1 as h → ∞, because exponent of h is negative.
1 1
(b) 1−q {1 − h1−q } → 1−q as h → 0, because exponent of h is positive.
√
4.(a) 1
s−2
, (b) 1
s+1
, (c) s2s−4 , (d) s
s2 +1
(e)1+s 3
2
2(s +1)
, (f) 1
s2 −4s+5
, (g) 4s4(s+1)
2 +8s+5 ,

2 6 2(s−1)
(h) s3
, (i) s4
+ s63 + s32 + 1s , (j) 1
(s+1)2
, 1
(k) (s−2)3 − 1
(s+2)3
, (l) (s2 −2s+2)2 .

s b
5. and . (Minus sign in denominator.)
s2 − b2 s2 − b2
s−1 1
6. (a) ∞
3
n=0 s = 1−s −1 = s−1 .
−n−1

s−2 1
(b) n=0 (−1)n s−2n−2 = 1+s
3∞
−2 = s2 +1 .

(c)* n=0 (−1)n s2n+1 = arctan(s−1 ).

3∞ −2n−1

" (T
7. (a) L f (t) = 0 te−st dt = −T es + 1−e−sT
! −sT
s2
,
" (π
(b) L g(t) = 0 e−st sin t dt = 1+e
! −sπ
s2 +1
.

1 1 1 1
8. LHS = L eat =
! "
s−a
, using tables. F (s) = s−1
from tables, so RHS = a (s/a)−1
.

(∞ (∞ 1
9. lims→0 0 e−st f (t) dt = 0 e0 f (t) dt. If f (t) = e−t , then LHS = lims→0 s+1 and
( ∞ −t
RHS = 0 e dt.
! " (T
If f (t) = cos t, then lims→0 L cos t = 0 but 0 cos t dt = sin T , which does not tend
to a limit as T → ∞.

10. No inverse transform: (a), (c) because F (s) ̸→ 0 as s → ∞.

11. (a) sinh t, (b) 12 (e2t − 1), (c) 2 − 2 cos 2t, (d) 1 − e−t (cos t + sin t),
(e) 41 {et (t − 1) + e−t (t + 1)}, (f) et − 1 − t, (g) t − 21 sin 2t, (h) 12 t2 − t + 1 − e−t .

b s
! "
12. LHS = L −b sin bt) = −b s2 +b2 . RHS = sL(cos bt)−cos 0 = s s2 +b2 −1 = LHS.

15
13. L(n!) = sn L(tn ) − 0 − · · · − 0 and L(n!) = n! 1s .

14. (a) y = t − 1 + (y(0) + 1)e−t ,

! "
(b) y = t + y(0) e−t ,
(c) y = −2t + 21 y(0) + y ′(0) + 2 et + 12 y(0) − y ′ (0) − 2 e−t .
! " ! "

(Note that D-operator methods are better for these simple examples.)
Particular solutions of (c): y = −2t + 23 et − 12 e−t and y = −2t + 2 sinh
sinh t
1
.

15. y(t) = 1 − e−2t cos t − 2e−2t sin t.

16.
(1 + 2s)Y + (3 + s)Z = (1 + 5s)/s2
(3 + s)Y − (1 + 2s)Z = 3/s2
leading to y = t + e−t sin t, z = 1 − e−t cos t.

! "
17. lims→∞ sF (s) = lims→∞ L f ′ (t) + f (0) = 0 + f (0).
! " (∞
Next lims→0 sF (s) = lims→0 L f ′ (t) + f (0) = 0 f ′ (t) dt + f (0) (by Question 9),
4 5∞
which equals f (t) 0 + f (0).

dn 1 n!
18. L tn = (−1)n ds = (−1)n (−1)(−2) . . . (−n)s−n−1 =
! " ! "
n s sn+1
.

d 1
19. tf (t) = sin t, so − ds F (s) = s2 +1
.

d 1 1
20. − ds F (s) = s+b − s+a . Inverting gives tf (t) = e−bt −e−at , so f (t) = 1t (e−bt −e−at ).
L−1 ln(1 + s−2 ) = 2t (1 − cos t).
1 2

d
(LHS) = −L t f (t)
! "
22. ds t
= −F (s), and LHS → 0 as s → ∞ since it is a Laplace
d d
(s
transform. Next, ds (RHS) = − ds F (p) dp = −F (s) (Fundamental Theorem of
( ∞∞
Calculus), and lims→∞ (RHS) = ∞ F (p) dp = 0.

23. (i) y = t + 14 t2 + 2. (ii) y = 21 t + t3 .

d d
24. − ds (s2 Y − sy(0) − y ′(0)) + (sY − y(0)) − ds Y = 0. Simplify and integrate.
Finally, by Question 17, y(0) = lims→∞ sY (s) = A.

16
25. LHS = lims→0 L f (t)
! " (∞ (∞
t
= lims→0 s F (p) dp = 0 F (p) dp = RHS (substitute
(∞ 5∞
s = p). If f (t) = sin t, then RHS = 0 s21+1 ds = arctan s 0 .
4

4(s2 +16) 16 s
26. F (s) = = −1 + 3s
s3 (s2 +12) √ 9s 2 + 9(s2 +12)
.
−1 8t2 cos 12t
f (t) = 9
+ 3 + 9 .

! n+1 "
28. If f (t) = tn then LHS = L tn+1 and RHS = 1 n!
s sn+1
. If f (t) = sin bt then
LHS = L 1b (1 − cos bt) and RHS = 1s s2 +b
b
!
2.

29. LHS = L et − t − 1) and RHS = s12 s−1

! ! "! 1 "
.

17
CHAPTER 4: TUTORIAL (Questions marked with an asterisk are for enrich-
ment.)
1. Prove the formula for f (t){H(t − a) − H(t − b)} given above. (Hint: evaluate
both Heaviside functions in each of the cases t ≤ a, a < t ≤ b, t > b, and then
simplify.)

2. Sketch the graphs of the following functions, and express each function by a single
formula, using Heaviside functions. ⎧
⎨0 for 0 ≤ t < 1
⎧ ⎪
⎪
⎨2 − t for 0 ≤ t ≤ 1 ⎪
(a) f (t) = (b) f (t) = t for 1 ≤ t < 2
⎩t for t ≥ 1 ⎪
⎪
⎩1 for t > 2
⎪
⎧ ⎧
⎨1 for 2n − 2 < t < 2n − 1 ⎨0 for 0 < t < 1
(c) f (t) = (d) f (t) = .
⎩0 for 2n − 1 < t < 2n ⎩ 1 for n ≤ t < n + 1
n
(In (c) and (d) the definitions apply for all natural numbers n: use sigma notation.)

! "
3. Use the formula for L f (t − a)H(t − a) to find the Laplace transforms of the
functions in Question 2.

e−s e−3s
4. Find the inverse Laplace transforms of 2 and 2 .
s −1 s − 2s + 5
⎧
⎨0 for 0 < x < a
⎪
⎪
⎪
5. Given y(0) = y ′ (0) = 0, solve the equation y ′′ − 3y ′ + 2y = 1 for a < x < b
⎪
⎪
⎩0 for x > b.
⎪
6. Give an alternative proof of the formula for the Laplace transform of a periodic
function by writing
+ ∞ ∞ +
7 (n+1)T
−st
e f (t) dt = e−st f (t) dt,
0 n=0 nT

(which is true for any function), substituting u = t − nT in the integral, and then
using periodicity.

7. Verify the formula for the Laplace transform of a periodic function with f (t) = eit .

18
8. Find the Laplace transform of:
i) The meander function
f (t)
✻

✲ t
a 2a 3a 4a

−1

ii) y = | sin t|, the “full wave rectification” of sin t.

9. Sketch the graphs and find the Laplace transforms of the periodic functions:
i) f (t) = t − 1 for 0 ≤ t < 2 and f (t) = f (t + 2) for all t > 0. (Saw-tooth wave.)
ii) g(t) = |f (t)|, where f (t) is as in i). (Full-wave rectification of f (t).)
iii) h(t) = f (t) where f (t) ≥ 0, h(t) = 0 otherwise, f (t) is as in i). (Half-wave
rectification of f (t).)

10. Repeat Question 9 i)–iii) with the square wave f (t) = A for 0 ≤ t ≤ T ,
f (t) = −A for T ≤ t < 2T , and f (t) = f (t + 2T ) for all t > 0. (For ii) and iii)
define g(t) and h(t) to be the full-wave and half-wave rectifications of f (t), as before.)

11. Find L(y) if ÿ + ω 2 y = h(t), where h(t) is as in Question 10 iii), and y(0) =
y ′ (0) = 0.

! "
*12. Show that if F (s) = L f (t) , then
∞
−1
! 1 " 7
L F (s) = f (t − nT )H(t − nT ).
1 − e−sT n=0

19
(Hint: expand (1 − e−sT )−1 as a geometric series. Note that f (t) is not necessarily
periodic.)
13. Solve y ′′ + y = δ(t − 2π) subject to:
i) y(0) = 1, y ′(0) = 0,
ii) y(0) = 0, y ′(0) = 0.

14. Show that if x(0) = ẋ(0) = 0, and if c2 − k 2 = ω 2 > 0, then the solution of the
equation ẍ + 2k ẋ + c2 x = m1 δ(t) is x = mω
1 −kt
e sin ωt.

15. If a > 0, prove that δ(t − a) ∗ f (t) = f (t − a)H(t − a) for a general function f (t)
by using the definition of convolution, considering the cases 0 < t < a and t > a
separately.

16*. Suppose y = z(t) is the zero-state solution of the D-operator equation P (D)y =
δ(t). It is called the impulse response of the system. Show that the zero-state
solution of P (D)y = f (t) is y = z(t) ∗ f (t), i.e., it is the convolution of the impulse
response and the input.

17. (a) Give a second proof of the formula Γ(α + 1) = αΓ(α) for α > 0 by writ-
ing tα = t(tα−1 ) and using the formulae for L tα , L tα−1 , and L t f (t) with
! " ! " ! "

f (t) = tα−1 .
(b) Give a third proof by using Chapter 2 Question 25 with f (t) = e−t tα . (Hint: show
that F (s) = Γ(α + 1)(s + 1)−α−1 .)

(∞
18. By substituting u = x1/α , where α > 0, show that 0
exp(−x1/α ) dx = Γ(1 + α).

19*. By considering the behaviour of Γ(α) as α → 0 and α → ∞, test for convergence

the improper integrals
+ 1 + ∞
1
Γ(α) dα and dα.
0 1 Γ(α)

20
 20. By substituting u = sin2 θ (as in Corollary 6), show that
+ π/2
B(α, β) = 2 sin2α−1 θ cos2β−1 θ dθ.
0
1 1 1
By putting α = 2
n + 2
and β =
and vice versa, deduce that
2
√
π/2 π/2
π Γ( 21 n + 12 )
+ +
n n
sin θ dθ = cos θ dθ = .
0 0 2Γ( 12 n + 1)
√
21. Show, by induction or otherwise, that Γ(k + 21 ) = (k − 12 )(k − 32 ) . . . 32 12 π for
k = 1, 2, . . . .
By writing out (2k)! in full, and taking 2 out of each factor, considering odd and
√
even factors separately, use the above to show that π(2k)! = 22k k!Γ(k + 21 ).

22*. By substituting u =+(1 + e−x )−1 in the integral definition of the Beta function,
∞
eαx
show that B(α, 1 − α) = dx for 0 < α < 1. This integral can be evalu-
−∞ 1 + e
−x
π
ated by complex variable techniques and shown to be equal to .
sin πα

ANSWERS TO CHAPTER 4 TUTORIAL

2. (a) f (t) = (2 − t)H(t) + 2(t − 1)H(t − 1).(b) f (t) = tH(t − 1) − (t − 1)H(t − 2).
∞ ∞
(−1)k H(t − k)
3 3
(c) f (t) = {H(t − 2n + 2) − H(t − 2n + 1} =
n=1 k=0
∞
3 1 3∞ 1
(d) f (t) = {H(t − n) − H(t − n − 1)} = H(t − 1) − H(t − n).
n=1 n n=2 n(n − 1)

5 5 5 5
$
4 $ 4 4 4
$
3 $ 3 3 3
2 $ 2 2 2
❅ $
1 ❅$ 1 $ 1 1

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

2 1 1 −s 1
3. (a) + 2 (−1 + 2e−s ) (b) (e − e−2s ) + 2 (e−s − e−2s )
s s s s
∞ 1 1 1 −s 3 ∞ 1
(−1)k e−ks =
3
(c) (d) {e − e−ns }.
k=0 s s(1 + e−s ) s n=2 n(n − 1)

21
 1
4. sinh(t − 1)H(t − 1) and H(t − 3)et−3 sin 2(t − 3).
2

5. 21 H(x − a){1 − 2ex−a + e2(x−a) } − 21 H(x − b){1 − 2ex−b + e2(x−b) }.

+ (n+1)T + T + T
−st −s(u+nT ) −snT
6. e f (t) dt = e f (u + nT ) du = e e−su f (u) du,
nT 0 0

using periodicity. Now sum from n = 0 to infinity, and use the formula for a geo-
metric series.
2π
1 1 1 4 (−s+i)t 52π 1
+
7. e(−s+i)t dt = e 0
= .
1 − e−2πs 0 1−e−2πs −s + i s−i
−as 2
(1 − e ) 1 − e−as coth(πs/2)
8. i) = , ii) .
s(1 − e−2as ) s(1 + e−as ) s2 + 1
1 1 1 + e−2s 1 1 1 1 1 − e−s
9. (i) − = − coth s. (ii) − 2
s2 s 1 − e−2s s2 s s s 1 + e−s
1 e−2s 1 e−s
(iii) − + .
s 1 − e−2s s2 1 + e−s

2 2 2
1 1 1
$ $ $ ❅ $❅ $❅ $ $
0 $ $ $ 0 ❅$ ❅$ ❅ 0 $ $
−1 $ 1 2$ 3 4$ 5 −1 1 2 3 4 5 −1 1 2 3 4 5
−2 −2 −2

A 1 − e−sT A A A 1
10. (i) = tanh(sT /2) (ii) (iii) .
s 1+e −sT s s s 1 + e−sT
1 A 1
11. .
s22
+ ω s 1 + e−sT
∞
!3 "
12. LHS =L−1 e−nsT F (s) .
n=0

13. i) ⎧
⎨cos t when 0 ≤ t ≤ 2π
y(t) =
⎩cos t + sin t when t > 2π .

22
 2 2 2
1 1 1
0 0 0
−1 1 2 3 4 5 −1 1 2 3 4 5 −1 1 2 3 4 5
−2 −2 −2

ii) ⎧
⎨0 when 0 ≤ t ≤ 2π
y(t) =
⎩sin t when t > 2π .

1 1 1 1
14. X(s) = 2 2
= .
m s + 2ks + c m (s + k)2 + ω 2
+ t
15. LHS = δ(u − a)f (t − u) du. If 0 < t < a, then the delta function is zero
0
throughout the interval of integration, so the integral is zero. If t > a, then the
integral is equal to f (t − u) evaluated at u = a, i. e., f (t − a). RHS equals zero if
t < a and equals f (t − a) if t > a.

16. Since P (D)z(t) = δ(t) we have P (s)Z(s) = 1. If P (D)y(t) = f (t), then

P (s)Y (s) = F (s), so Y (s) = P 1(s) F (s) = H(s)F (s). Now invert and use the convo-
lution formula. (This shows that the transfer function P 1(s) is equal to the impulse
response Z(s).)

Γ(α + 1) d Γ(α)
17. (a) From formulae we get = − . Simplify the RHS and cancel.
+ ∞ sα+1 ds sα
Γ(α + 1)
(b) Γ(α) = ds, which can be evaluated.
0 (s + 1)α+1

18. Put u = x1/α , so x = uα and dx = αuα−1 du.

1
1
+
1
19. Divergent, since Γ(α) ≈ α
for α near 0, and dα diverges.
0+ α
∞
1
Convergent, since 0 < Γ(α)
< e−α for large α, and e−α dα converges.
T
(Γ(α) grows faster than exponentially.)

23
20. Note that 1 − u+= cos2 θ and du = 2 sin θ cos θ dθ.
π/2
1 1 1 1
In the second part sinn θ dθ = B( n + , ).
0 2 2 2 2

21. (2k)! = {(2k)(2k − 2) . . . (4)(2)}{(2k − 1)(2k − 3) . . . (3)(1)}

= 2k {k!}2k {(k − 21 )(k − 32 ) . . . 32 12 }.
Now substitute.

24
CHAPTER 5: TUTORIAL (Questions marked with an asterisk are for enrich-
ment.)

1. Prove the properties i) to v) in Corollary 8.

2. If f (t) has period 2 and f (t) = 1 − t for −1 < t < 1, find the Fourier series of
f (t). Sketch the graph of the function to which the series converges between t = −4
and t = 4.

3. Find the Fourier series of the function f (x) with period 1 such that
⎧
⎨0, for − 12 < x < 0
f (x) =
⎩x, for 0 < x < 12 .

Sketch the graph to which the series converges between x = 0 and x = 4.

By considering the value to which the series converges when x = 12 , show that

1 1 1 π2
1+ + + +··· = .
9 25 49 8

4. Find the Fourier series of the function f (x) of period 2π such that f (x) = x2 for
−π < x < π.
By considering x = 0 and x = π, deduce that
∞ ∞
7 (−1)n−1 π2 7 1 π2
= and = .
n=1
n2 12 n=1
n2 6

ANSWERS TO CHAPTER 5 TUTORIAL

2 3∞ (−1)n
2. 1 + sin nπt .
π n=1 n
(−1)n+1 1 − (−1)n
) *
1 1 3 ∞
3. + sin 2nπx − cos 2nπx .
8 2 n=1 nπ n2 π 2

25
 1
Converges to 4
at x = 12 , (jumps from 1
2
to 0), so
∞ )
1 − (−1)n
* ) *
1 1 17 n 1 1 1 1
= + 0− (−1) = + 2 1 + 2 + 2 + · · · .
4 8 2 n=1 n2 π 2 8 π 3 5

∞ (−1)n
π2
cos nx. Converges to 0 at x = 0 and to π 2 at x = π. (Points of
3
4. 3
+4 2
n=1 n
discontinuity).

26