MANUAL FOR
ZILL'S
A FIRST COURSE IN
DIFFERENTIAL
EQUATIONS
WITH MODELlNG ApPLlCATIONS
7TH EOITION
ANO
ZILL& CULLEN'S
DIFFERENTIAL
EQUATIONS
WITH BOUNDARyVALUE PROBLEMS
5TH EOITION
BROOKS/COLE
THOMSON
LEARNING
Table of Contents
1
22
71
104
194
240
308
370
430
10
458
11
491
12
538
13
616
14
654
15
695
Gamma function
717
Introduction to Matrices
718
Appendix 1
Appendix II
1. Secondorder ; linear.
+ y2 = 1, we see that it is
1) (dx / dy) + x = O, we see that
+ (1 + u)v =
ue'U we see
+ u = O, we see that it
is nonlinear in u.
5. Fourthorder; linear
6. Secondorder; nonlinear because of cos(r
7. Secondorder; nonlinear because of
+ u)
JI + (dy/dx)2
12. Frorn y
&
13. From y
+ 20y =
=
y" = tanx
+ 13y
+y
= _Cx/2
+ ex/2
= O'.
= 24e20t, so that
24e20t
+ 20 (~
 ~e20t)
= 24.
= 3e3x cos 2x  2e3x sin 2x and y" = 5e3T cos 2x  12e3T sin 2x, so
= O.
+ tan x)
ln(secx + tanx).
cos x ln(sec x
+ cosx
Then 2y'
Then y"
+ Y = tanx.
1
+ tan x)
and
Exercises
1. 1
=t
nd differentiating implicitly .
we obtain
2
dX
2X  1 dt
1 dX
X  1 dt
=1
( 2X 2
1)
dX
dt
1 X  1
~4~2~\~2~4t
=1
2X  2  2X + 1 dX
=1
(2X  l)(X  1) dt
dX
dt
2
4
= (2X
 l)(X
 1) = (X  1)(1  2X).

X1
Solving et
= et
==:> 2X  1
O we get t
= Xet
et ==:> (et
 1)
= (et  2)X
==:> X
et  1
t.
e2
or on (ln2, (0).
The graph of the solution defined on (00, in 2) is dashed, and the graph of the solution defined on
(ln 2, (0) is solid.
16. Implicitly differentiating the solution we obtain
2 dy
 2x   4xy
dx
+ 2y dy
dx
= O==:>
x dy  2xy dx
==:> 2xy dx
+ (x2
+ y dy
=O
 y)dy = O.
2
4
y = (2x2
solutions
= el et /
(1 + el et) we obtain
dP
(1 + elet) elet
dt
= ~~,,
elet. elet
(1 + elet)2
Exercses
Substituting into the differential equation, we have
2 {X
2
y' + 2xy = 1  2xeX Jo et dt  2clxex
19. Fromy
= cle 2x+c2xe
weobtain dy
dx
+ 2xeX
?
= (2cl+c2)ex+2c2xex
(X 2
Jo el dt + 2clxex
d2y
and 0
dx
1. 1
= l.
= (4Cl+4C2) e 2x +4C2xe 2x ,
so that
20. Frorn y
= qxl
dy = C1X
dx
+ c2 + c3 + c31n x +
8x,
d2y
3
1
dx2 = 2CIX + C3X + 8,
and
so that
3 d3y
x dx3
+ 2x
d2y
dy
dx2  x dx
+y
+ (C3 + c3)xlnx
so
 2(x2)
= O.
Thus, rPl and rP2are solutions of the differential equation on (00, (0).
(b) Frorn y
x2
2'
{ z,
x <O
{2X.
we obtain y' =
'
x ;:: O
2x,
is not continuous at x
= Osince x~olim
x <O
so that xy'  2y = O.
x ;:: O
y(x)
= 5 and
lim y(x)
x~o+
= O.
Exercises
(b) From
1. 1
= 1 + (x
+ 3)1/2 we have .
(y  x)y'  y + X  2 = [x + 2VX+3  x][l + (1 + (x  3)1/2]
y!
 [X + 2Vx+3] + x  2
= 2vX+3 + 2  x  2Vx+3 + x  2 = O.
(b) For (1 sin t)l/2 to be continuous we must have 1 sin t > Oor sin t < 1. Thus, (1 sin t)l/2
will be a solution on (K/2, 5K/2).
= memt.
Then y'
+ 2y =
O implies
+ 2emt = (m + 2)emt = O.
memt
m2emt _ 5memt
+ 6emt = (m 
2)(m  3)emt = O.
Since emt > O for all t, m = 2 and m = 3. Thus y = e2t and y = e3t are solutions.
26. (a) From y
= t'" we obtain y' = mtm1 and y" = m(m  1)tm2 Then ty" + 2y' = Oimplies
tm(m  1)tm2 + 2mtm1 = [m(m  1) + 2m]tml = (m2 + m)tm1
= m(m + 1)tm1
= O and
1. Thus y
= m(m
= O.
1 and 7J =
 1)tm2
el
are solutions.
+ 15y =
implies
t2m(m
+ 15tm =
 1)tm2  7tmtm1
[m(m  1)  7m
= (m2
 8m
+ 15]tm
+ 15)tm
= (m  3)(m  5)tm
Since t'" > O for t > O, m = 3 and m = 5. Thus y = t3 and y = t5 are solutions.
27. From x = e2t + 3t and y = _e2t + 5e6t we obtain
dx
dt
_2e2t
+ 18e6t
and
dy = 2e2t
dt
+ 30e6t
Then
x
+ 18e6t =
and
dx
dt
= O.
Exercises
+ 3y
5x
28. From x
= 5(e2t
=
1. 1
+ 30e6t
2e2t
= dy .
dt
and
d2x
dt
2
1t
+ 2 cos 2t +  e
4cos2t
1
5
+ _et
 4sin2t
dy.
1t
and
and
2
d2y
= 4 cos 2t
dt
+ 4 Sll12t   el .
5
Then
4y
4 cos 2t  4 sin 2t
+ Se
d2x
dt2
and
4x  et
= 4(cos 2t
+ sin 2t + ~et)  et
d2y
dt2 .
1 t
= 4 cos 2t + 4 sin 2t  Se
29. (y')2
f(t)
cet or y
f(t)
respectively.
c sin t or y
derivatives that are the negatives of themselves. The differential equation is y"
The corresponding
c cos t have second
+ Y = O.
33. Since the nth derivative of 4>(x) must exist if 4>(x) is a solution of the nth order differential equation,
alllowerorder derivatives of 4>(x) must exist and be continuous. [Recall that a differentiable function
is continuous.]
34. Solving the system
for
and
C2
Cy(O)
+ C2Y2(0) = 2
Cy~ (O)
+ C2Y2(0)
=O
we get
and
C2
2y~ (O)
=  _,.:.~'::_y (0)Y2(0) 
yi (0)Y2(0)
Exercises
1. 1
+ clt + C2
36. Solving for yl using the quadratic formula we obtain the two differential equations
= f(t,y).
= t. These are not equivalent because
y = O is a solution of the first differential equation but not a solution of the second.
38. Differentiating we get yl = CI
+ 3C2t2
Y 
+ 3y
C2
= ty I 
el
= yl  ty" /2, so
1 2 11
3't y
= O.
(d) Setting yl = O we have V" = O and 6y = 10. Thus V = 5/3 is a constant solution.
41. One solution is given by the upper portion of the graph with domain approximately (0,2.6). The
other solution is given by the lower portion of the graph, also with dornain approximately (0,2.6).
42. One solution, with dornain approximately (00,1.6)
quadrant together with the lower part of the graph in the first quadrant.
domain approximately (0,1.6) is the upper part of the graph in the first quadrant.
solution, with domain (0,00), is the part of the graph in the fourth quadrant.
The third
Exercises
+ y3)/xy = 3c w
xy(3x2 + 3y2y') _
:;J'?J
3x3
obtain
t;
1. 1
x3
+ y3)(xy' + y)
=O
+ 3xy3y' 9'x3_y_:xy3y'
1:;:;,y3 _ x4
_ y4
xy3)y'
=O
= 3x3y
+ x3y + y4
44. A tangent line will be vertical whor y' is undefined, or in this case, where x(2y3  x3) = O. This
gives x
+ ~x3
= 3x
C;/3
3 3
3
x = x
2
21/3
x)
x3 = 22/3x2
Thus, there are vertical tangent Iines at x = O and x = 22/3, or at (O,O) and (22/3,21/3) _ Since
22/3 :::::1.59, the estimates of the dornains in Problem 42 were close,
45. Since q;'(X) > O for all x in J, q;(x) is an increasing function on J. Hence, it can have no relative
extrema on J.
46. (a) When y
(b) When y
solution must be increasing. Thus, none of the curves in color can be solutions.
(c)
5
y(a  by)
> O [01' O < y < afb, y = q;(x) is increasing on this interval. Since
.11
= q;(x)
Exercises 1. 1
(c) Using implicit differentiation we compute
~:; = y( by')
+ y'(a
= Owe obtain y = a/2b. Since d2y / dx2 > Ofor O < y < a/2b and d2y / dx2 < O
< y < a/b, the graph of y = cp(x) has a point of infiection at y = a/2b.
C2
Cl
1, 4, lO. The
use
Clear[y]
y[x_]:= x Exp[5x] Cos[2x]
y[x]
Y"" [x]  20 Y'" [x]
50. In Mathematica
use
Clear[y]
y[x_]:= 20 Cos[5 Log[x]]/x  3 Sin[5 Log[x]]/x
y[x]
x3 Y'" [x]
Exercses
1.2
Exercises 1.2
1
1. Solving 3 =

1 + Cl
_
1
1  4et
.11
2. Solvmg 2 = 
Xl
5. Using x'
= cost + 8sint.
The solution is x =  cos t.
J3 + C2
1
1
= 2
2
Cl
2
1
Cl
2
.
S o1vmg
we fin d Cl =
6. Using x'
J3
4
J3 = O.
+ C2
2
IS
x =
J3 cos t + 4:l. sm t .
4
V2
Cl
2
V2
= V2
+C2
2
 
~ and C2= 
~ex  ~ex.
= ~ and
C2=
 ~e2
= e.
 ~e2x.
Clel 
C2e =
5.
problern is y
= 5exl.
Exercises
1.2
Cl 
Solving we get
Cl
= C2 = O. A solution
= Oand y = x3
= Oand y = x2
C2
= O.
= O.
EL
solution.)
13. Por f(x, y) = y2/3 we have ~f = ~yl/3 Thus the differential equation will have a unique solution
uy
3
in any rectangular region of the plane where y
i= o.
14. Por f(x, y) = .xy we have ~f = ~ ~. Thus the differential equation will have a unique solution
2y y
uy
in any region where x> Oand y > Oor where x < O and y < O.
15. Por f(x, y) = Ji. we have ~f = ~. Thus the differential equation will have a unique solution in any
x
uy
x
region where x i= o.
af
16. Por f(x, y) = x + y we have ay = l. Thus the differential equation will have a unique solution in
the entire planeo
17. Por f(x, y)
2 2
af
X Y '). Thus the differential equation will have a unique
we have ;:; =
uy
(4  y2
x2
2
4 y
x2
af
_3x2y2
we have ;:; =
2' Thus the differential equation will have a unique
uy
(1 + y3)
3
1+y
i=
l.
y2
af
2x2y
we have ;:; =
2' Thus the differental equation will have a unique
x +y
uy
(x2 + y2)
solution in any region not containing (O,O).
= 
y +x
8f
2x
we have ;:; = (
)2'
yx
uy
yx
10
Exercises
1.2
24. The differential equation is not guaranteed to have a unique solution at (1,1).
25. (a) A oneparameter family of solutions is y = ex. Since y' = e, xy' = xe = y and y(O) = e 0= O.
(b) Writing the equation in the form y' = y/x we see that R cannot contain any point on the yaxis.
Thus, any rectangular region disjoint from the yaxis and containing (xo, Yo) will determine an
interval around Xo and a unique solution through (xo, yo). Since zo = O in part (a) we are not
guaranteed a unique solution through (O,O).
(e) The piecewisedefined function which satisfies y(O)
entiable at x = O.
d
+ e) =
sec2(x
+ e)
= 1 + tan2(x
+ e),
+ e) satisfies
the
differential equation.
(b) Solving y(O) = tan e = O we obtain e = O and y = tan x.
x = 7l' /2, the solution is not defined on (2,2) because it contains 7l' /2.
(e) The largest interval on which the solution can exist is (7l' /2, 7l'/2).
27. (a) Since
!!_ ( __
dt
equation.
1_)
+e
( t+e )
t+c
= Y , we see that y =
+ t).
= 1
(1,00).
(e) Solving y(O)
= l/e
= Yo we obtain e = l/yo
1
Y =  l/yo
+t =
and
Yo
1 _ yo t '
yO =1 O.
+ t =1 O, the
when yo
= Ois a
solution on (00,00).
Exercises
1.2
= 3 for y
Y = <P2(X) = J3(x2
= <P4(X) = J3(x2
< 00,
oo<x<l,
4
 1) ,
= 2 and y
< 00,
1< x
 1) ,
 1) ,
Y = <P3(X) = J3(x2
(e) Setting X
1< x
1) ,
y = <Pl(X) = J3(x2
we get
00
= 3.
= 4 in 3x2 _y2
= e we get 1216 = 4 = e,
+4,
00
4
2
4
V3x, both
defined on (00,00).
1, so the only plausible solution curve is the one with negative slope
= dyjdx
f(x,y)
is determined by f(x,y),
= xo
31. If the solution is tangent to the xaxis at (xo, O), then y'
these values into y'
+ 2y
=O
when x
= Xo
and y
= O.
Substituting
= 3x  6 we get O + 0= 3xo  6 or Xo = 2.
32. We look for the curve containing the point corresponding to the initial conditions.
(a) D
(b) A
(e)
(d)
(e) B
(f) A
33. The theorem guarantees a unique (meaning single) solution through any point. Thus, there cannot
be two distinct solutions through any point.
34. The functions are the same on the interval (0,4), which is all that is required by Theorem 1.1.
12
Exercises
=x
O, for y
1.3
y' > O, for y > x  1, we see that solutions aboye the line y
are increasing. From y' < O, for y < x 
1,
=x  1
x, we see
Exercises 1.3
dP
dP
= kP+r'
 = kPr
dt
'
dt
2. Let b be the rate ofbirths and d the rate of deaths. Then b = k1P and d = k2P. Since d.Pf dt = bd,
1. 
= kl P  k2P2.
5. From the graph we estimate To = 180 and Tm = 75. We observe that when T = 85, dT/dt ~ 1.
Frorn the differential equation we then have
k = dT/dt
TTm
1
8575
= 0.1.
~ =
k [T 
(80  30 cos ;2
t) ], t > O.
7. The number of students with the flu is x and the number not infected is 1000  x, so dxf dt. =
kx(1000  x).
8. By analogy with differential equation modeling the spread of a disease we assurne that the rate at
which the technological innovation is adopted is proportional to the number of people who have
adopted the innovation and also to the number of people, y( t), who have not yet adopted it. If one
person who has adopted the innovation is introduced into the population then x
dx
dt = kx(n + 1  x), x(O) = 1.
13.
+y
=n
+ 1 and
Exercises
1.3
C~O
lb/gal) = 1~ lb/mino
1 gal/min. After t minutes there are 300 + t gallons of brine in the tank. The rate at which salt is
leaving is
R2
= (2 gal/min) (  A) lb/gal
300 + t
2A
lb/mino
300 + t
_1 dV
Aw dt
= Awh.
_1 (cAofiih)
= _ cAo fiih.
Aw
Aw
450
Problem 11 for the volume of water leaving the tank we see that the differential equation is
dh
dt
Using Ah
= 7r(2/12)2 = 7r/36,
=  3cAh
Aw
Aw
fiih.
dh = _ 3(0.6)7r/36
dt
Aw
= 7r(2h/5f
= 47rh2/25. Thus
dq
+ R dt
= E(t) we obtain
di
L dt + Ri = E(t).
14
8/20, so r
2h/5 and
Exercises
dv
= kv
dt
1.3
+ mg.
= (62.4)7T(S/2)2y = 15.67Ts2y.
w d2y
2
9 dt
15.67Ts 'Y
d2x
F = ma = m dt2 = k(s
Since the condition of equilibrium is mg
+ x) + mg
+ mg
= kx
 ks.
= +ks.
m dt2
18. From Problem 17, without a damping force, the differential equation is md2x/dt2
= kx.
With a
dx
d2x
= kx  (3 dt
or
m dt2
dx
+ (3 dt + kx = O.
19. Let x(t) denote the height of the top of the chain at time t with the positive direction upward. The
weight of the portion of chain off the ground is W = (x ft) . (1 lb/ft) = x. The mass of the chain is
m = W/g = x/32. The net force is F = 5  W = 5  x. By Newton's second law,
d x
dt (32 v) = 5  x
dv
x dt
or
dx
+ v dt
= 160  32x.
+ (dx)2
+ 32x =
dt
160.
d
20. The force is the weight of the chain, 2L, so by Newton's second law, d [mv] = 2L. Since the mass
!!_ 2(L
dt
 x) v] = 2L
9
(Lx)+vdv
dt
or
15
= Lg.
(dX)
dt
=Lg.
Exercises
21. From g
1.3
kj R 2 we find k
gR~.0 Using a
upward we get
or
22. The gravitational force on mis F = klVIrmjr2.
u, = r3 M j R3
Since NIr
= 41fr3j3 and M
= 41fR3j3 we have
and
F = k Mrm = k r3VImjR3 = k mM r.
r2
Now from F
r2
R3
dt
dA
24. The differential equation is  = kl(M  A)  k2A.
dt
25. The differential equation is x'(t) = r  kx(t) where k > O.
26. By the Pythagorean Theorem the slope of the tangent line is y'
27. We see from the figure that 28
y
x
= tan
+ o: = tt,
o: = tan(7T 28)
y
r====
Js2  y2
Thus
=
tan 28 =
2 tan 8
1 _ tan2 8 .
Y =
2x
J4x2
2y
+ 4y2
x
Jx2
+ y2
.x
 x
+ J x2 + y2
or
y ~~  Jx2
+ y2 + X
= O.
= k(T  Tm).
16
Exercses
> O and
1.3
= 6  A/I00.
A'(t)
> O
input rate of salt is riCi lb/mino The output rate of brine is ro gal/min and the concentration of
salt in the outfiow is colb/gal, so the output rate of salt is rocolb/min.
Vo + (ri  r o)t gallons of brine in the tank, and the output rate of salt is r oA/[Vo
+ (ri 
ro)t]lb/min.
32. This differential equation could describe a population that undergoes periodic fiuctuations.
dP
33. (1):
di =
(3):
di =
(6):
dX
dt
(10):
dT
kP
is linear
(2):
(5):
dh
dt
=
Ah ~
dA
di
dx
dt
= kx(n + 1  x) is nonlinear
dA
A
=6 is linear
100
(8): 
dt
d2q
Aw V 2gh is nonlinear
(11):
is linear
= kA
L dt2
dq
+ R dt +
Cq
= E(t)
is linear
dv
(14): m = mg  kv is linear
(12):
dt
d2s
d2x
ds
= g
(16): 
dt2
64
 x = O is linear
L
+ s,
= d2s / dt2
+ s)
2
;:::;gR
[R
2
 2sR
3
+ ... ] = g +
2gs
R3
+ ....
Thus, for R much larger than s, the differential equation is approximated by d2s/dt2 = g.
35. If p is the mass density of the raindrop, then m = pV and
dm = p dV
di
dt
< O. Solving
dr/dt
dt
pS dr .
dt
+ ro.
17
Exercises 1.3
d
Frorn Newton's second law, [mv]
dt
dv
m dt
= mg, where
dm
or
+vdi = mg
3k
dt
pr
+v=g
dv
or
3k/ p
+ k t / P + ro v
d
= g,
< O.
36. We assume that the plow clears snow at a constant rate of k cubic miles per hour. Let t be the
time in hours after noon, x(t) the depth in miles of the snow at time t, and y(t) the distance the
plow has moved in t hours. Then dy/dt is the velocity of the plow and the assumption gives
dy
wx=k
dt
where w is the width of the plow. Each side of this equation simply represents the volume of snow
plowed in one hour. Now let to be the number of hours before noon when it started snowing and
let s be the constant rate in miles per hour at which x increases. Then for t
> to, x
= s(t
dy
dt
=
ws t + to
Integrating we obtain
k
y =  [In (t + to) + e]
ws
where e is a constant. Now when t = 0, y = so e = In to and
y = ~ In (1
ws
..!.)
.
to
(1 + ~)
Expanding and simplifying gives
ta + to 
= 3, we obtain
(1 + t~f
1 = O. Since to
18
+ to).
Chapter
1 Review Exercises
dy
dx
d
2. dx (5 + cle2x)
d
3. (q eos kx
dx
= 2cle2x
+ C2sin kx)
d2
2 (Cleos kx
dx
+ cle2x
= 2(5
= kCl sin kx
+ C2sin kx)
dy
dx
 5)
+ kC2 eos kx
d2y
d2y
= _k2y
or + k2y = O
dx2
dx2
d
4.  (Cleosh kx + C2sinh kx) = kc sinh kx
dx
+ C2sin kx)
+ C2sinh kx)
d22 (Cleosh kx
dx
d2y
2
dx2 = k y
5. y'
or
= k2 Clcosh kx
+ k2 C2sinh kx
= k2 (ci eosh kx
+ C2sinh kx)
d2y
2
dx2  k y = O
y"
+ kC2 eosh kx
y"
+ C2xex + 2C2ex;
= 2(C1eX + C2xex + C2eX) = 2y'
= C1ex
y"  2y'
+y
=O
8. e
y"  2y'
+ 2y
eos x
=O
10. a,e
9. b
= O, y
+ c2ex
C2ex sin x
11. b
12. a,b,d
= e, and y = eX.
y" +y'
+ y2.
y'
or
= O.
= x2/3 is undefined at x
+ Cwe obtain
19
1 Review Exercises
Chapter
JX2=X = 1 vx(x  1). Since x(x  1) ;:: O for x S O or x ;:: 1, we see that neither
=1
= 1+
vx(x  1) nor y
solutions of the differential equation, but neither is a solution of the initialvalue problem.
19. (a)
y
3
3
(b) When y
= x2 + el. y'
= 2x and (y')2
')
= 4x2. When y
{_x
6V4 +
5( 1)3
x<O
2,
2
, x S O, we get y =
21. Differentiating y
x,
.
= 7.
2 '"
x y + xy + y
= x 2( 
sin(lnx) + cos(lnx))
+x
cos(lnx)
. (1 )
+ sm n x
Then
= O.
= cos
(1)
nx
1
(Sin(lnx))
(1 ) cos nx
x
1 (
+ neos
(1
nx
))(
sin(lnx))
1 cos(lnx)
+ nx
x
x
+ sin(lnx)
x
(ln x) cos(ln x)
x
+~'''
and
y
,,[
1
(Sin(lnX))]
(1 ) cos nx
x
1
[
(Sin(lnx))
+ ln( cos(ln x)) sin(ln x)"2 + x (In x) x
x
2
1
"2
x
COS(lnX)] 1
?
x
z
1[
sin (ln x)
.
="2  ln( cos(ln x) cos(ln x) +
(1 ) + ln(cos(ln x) sm(ln x)
x
cos nx
 (lnx) sin(lnx) + cos(lnx)  (lnx) cos(lnx)].
20
1
(In x) cos(ln x)"2
x
1 Review Exercises
Chapter
Then
2
x 2 y 1/
+ XY / + y =
+ cos(ln x)
+ (In x)
=
In ( cos(lnx))cos(lnx)
sin2(lnx)
cos(ln x)
. ()lnx
sm
cos lnx
+ cos(ln x)
ln(cos(ln x))
+ (In x)
sin2(lnx) + cos2(ln x)
cos(ln x)
+ cos ()In x =
sin (In x)
1
cos(ln x)
Using Ao
= _ cAo
= O.
fi;h.
Aw
J64h
V3
= mg  {Jmg
=~
288
. ( lnx )
 (lnx ) sm
or
21
vh.
= sec (1)n x
1.
7.
2.
8.
\ I \ 1\1 \
....._'"'""
\ I
\ I
\ 1\ 1\ I \
\ I \ lit
\
\1 \ 1II \
\ \ ,""
\ \ , .._'"'"
\\"
..~,
\ I \
\ \"
I I I \
\ I
\ I
\ I \ I I I \
\ I \ I I I \
\ I
\I \ I I I \
.........
......... ~
~
.........
.........
22
..........
\ I
__
~,
~,
__
<lo ..
"
,
,~
,~
,~
,~
......
_
__
......
Exercises 2. 1
9.
10.
y
II I 11I 11I I
I I I
1111/111'
I
1111/1'
1/.
111"
1'1'
I
y
111
11
11
111
,.,
II
II1
II1
III
II
II
II
III
~~~~~.f+~~~~'~'
1X
1""
..... "
I ,
I , 1" I , .._"
I , l' I " .._,
l'
,\ I ",
...
11.
1\ 1, I ",
I I 11I '1
III
11
11
12.
y
II
I l'
"11/111111
"11/111111
I III
1"
J"
11""
13.
1I
"
I .. ,.,."
.,
II
11
11
,,,
'"
,,,
111
1111
1111
III
III
III
1111
111.
11
1111
111I
1111
1II1
'11111
1111"
II
, 1II
111.
1111'
1111'
1111
111I
.. ,,"
"
"
"
1\ 1 1 1\1
'1"
I
I
1 ........
,"
_
_

III
III
III
1'11
I1II
11II
I l' , , l'
1111'111'
I1I1
1111'
11II
111"
'"
1111 111
" I
"
I
11I
,"
"
""""'\
""
'"
.., 1
.. \  1
..\1
..\1
11
I
I
I I
I
II
," ,
"11
11 \
11
...
11
11 \11_ \_
\J
11 \11 ... ' _
II
II
,,
"III I
1'
I I
11 \
111111
11 \
11,
, , ,
I
I
I
"
II
1 1\
11
= 1, y = O,and y =
1'"
1""
y
_1
I1  \  I
I 1 \1
11' .
1 1'
11'
11'
"
I
I
, I
" "'\\'\,
"1
""
", I "'1 ""
 .. _ .. ""
, I r r r 1"
II 11 I 1'"
I
11 I
I 11 ,
II11 I 1""
1 .........
I
I
I
I
"""""
.. ........
"""""
X
,\ ""
,,,
"
,,,
1" , , 1" , , ,
,"
11""
11""
\,,,
\ "
.........
_ ......
\ \
 __ "".. "
....."
'"'_.._ ...
,'"
14.
y
1111
"
.. I
II
II
II
....
I "1
, I111
I ",,
'11111/111
'1//1
I I11I
l'
11
II
II
y
, 1'"
y)(l
(a)
(b)
(c)
(d)
23
+~~2x
Exercises 2. 1
16. Writing the differential equation in the form dyjdx
= y2(1  y)(l
+ y)
1
(a)
(b)
iL
2
1
2 x
(d)
(c)
2
3
.
5
From the phase portrait we see that Ois asymptotically stable and 3 is unstable.
18. Solving y2  y3
= y2(1  y)
o
From the phase portrait we see that 1 is asymptotically stable and Ois semistable.
19. Solving (y  2)4
2.
+ y) = Owe obtain
2
From the phase portrait we see that 5 is asymptotically stable and 2 is unstable.
21. Solving y2(4  y2)
= y2(2  y)(2 + y)
2
= Owe
24
Exercses 2. 1
From the phase portrait we see that 2 is asymptotically stable, Ois semistable, and 2 is unstable.
22. Solving y(2  y)(4  y) = Owe obtain the critical points 0,2, and 4.
From the phase portrait we see that 2 is asymptotically stable and O and 4 are unstab!e.
23. Solving y ln(y
+ 2) = Owe obtain
2
1
From the phase portrait we see that 1 is asymptotically stable and O is unstab!e.
24. Solving yeY
9y = y(eY
In 9
From the phase portrait we see that O is asymptotically stable and In 9 is unstable.
25. (a) Writing the differential equation in the form
dvdt = mk (mkg _ v)
we see that a critical point is mg j k.
rng/k
From the phase portrait we see that mg j k is an asymptotical!y stable critica! point. Thus,
limt>oov
= mg j
k.
Jmgjk.
J rng/k
J mg j k is an asymptotically
= Jmgjk.
26. (a) From the phase portrait we see that critica! points are a and (3. Lct X(O)
= Xo.
If Xo
t
a as t
t
25
r+
a as t
r+
oo.
Exercses 2. 1
If Xo
> (3, we see that X(t) increases in an unbounded manner, but more specific behavior of
X (t) as t
+ 00
is not known.
If Xo
+
This could happen in a finite amount of time. That is, the phase portrait does not indicate
that X becomes unbounded as t
+
oo.
= dt
=t+c
a X
aX=
t+c
X=a.
For X(O)
For X (O)
= a/2
t+c
we obtain
1
X(t)=a
/
t+2 a
X(t)=a
/
t 1a
2a we obtain
1
26
Exercises 2. 1
= c.
y
1111111111
1 r I I 11 I I I I
I 11 J 1I I
I JI
1 11 111
1 1 111 1 11 11
1111
11 1/ 1 11/1
III~IIIII
\""""
\ \ , \ \ \ I \,
""'"
'\\"1'"
\ \ , \ \ \ 1"
"""""
2.2,
,\
, I \ \ \ I \ \ \
, I \ \ \ I \ \ \
\1 1111111
\1 1111111
1111111111
1111111111
points are y
11
III
II
I 1I II
1"1
I111
/1/11/111/
111111111'
\
\111111\11
\ \ \ 1 \ \ 1 \ \ \
28. Critical
1 111111
\ \ \\ l' \\
""""
1111111111
111
I 11 I 11
III~IIIII
II111
I 11
111111111
11111111
I I
0.5, and
y = 1.7.
0.5
2.2
1.7
29. At each point on the circle of radius c the lineal elernent has slope c2.
30. (a) When x = Oor y = 4, dy/dx
dy/dx
When y
= 3 or y
= 5,
(b) At (O,YO) the solution curve is headed down. As x increases, it will eventually turn around and
head up, but it can never cross y
O. Thus, y cannot approach
31. When y
00
as x approaches oo.
> ~x2,
/'.1.
I I I
I , I
I I
I 1
I I
I I I
I I
I I I
J
J
I
JI'
I
III
l'
I
I I
I I
I I
I I
I
I I
3~~~~~~O~/~/~~~I~I~;~I~I~~~
32. For dxl dt = Oevery real nurnber is a critical point, and hence all critical points are nonisolated.
33. Recall that for dy / dx
f(y)
f'
sorne interval J. Now suppose that the graph of a nonconstant solution of the differential equation
27
Exercses 2. 1
crosses the line y
= c. If the point of intersection is taken .as an initial condition we have two distinct
solutions of the initialvalue problem. This violates uniqueness, so the graph of any nonconstant
solution must lie entirely on one side of any equilibrium solution. Snce
change signs around a point where it is O. But this is a critical point. Thus, f(y) is completely
positive or completely negative in each region R; If y(x) is oscillatory or has a relative extremum,
then it must have a horizontal tangent lne at sorne point (xo, Yo). In this case yO would be a critical
point of the difIerential equation, but we saw aboye that the graph of a nonconstant solution cannot
intersect the graph of the equilibrium solution y = yo.
34. By Problem 33, a solution y(x) of dyjdx
monotone. Since y'(x)
= f(y)
= f(y) > O, y(x) is monotone increasing, and since y(x) is bounded aboye
by C2, limx+ooy(x) = L, where L ::; C2. We want to show that L = C2. Since L is a horizontal
asymptote of y(x), limx+ooy'(x)
f(L)
= f( x+oo
lim y(x)) = Iim
xoo
Cl
f(y(x))
= lim
xoo
y'(x)
= O.
Cl
and
35. (a) Assuming the existence of the second derivative, points of infiection of y(x) occur where
y"(x) = O. Frorn dyjdx = g(y) we have d2yjdx2 = g'(y) dyjdx. Thus, the ycoordinate of a
point of infiection can be located by solving g' (y) = O. (Points where dy j dx = Ocorrespond to
constant solutions of the difIerential equation.)
(b) Solving y2_y_6
= (y3)(y+2)
(2y  1) dyjdx
i,
although the
>O
< y < ~ and y > 3, we see that solution curves are concave
down for y < 2 and i < y < 3 and concave up for 2 < y < i and
y > 3. Points of inflection of solutions of autonomous difIerential
s
for 2
5
equations will have the same ycoordinates because between critical points they are horizontal
translates of each other.
28
Exercises 2.2
Exercises 2.2
In many of the following problems
solve for g(y) we exponentiate
implies g(y)
ecef(x).
wc will encounter
Letting
3. Frorn dy
4. From (
el
dx we obtain In
Iyl =
1
6. From  dy = 2x dx we obtain In
y
41n
From
+ 2 + ;)
(y
11.
dy =
cscy
 cos y =  ~x 
ll. From 
2
sec x
+ 5)2 dx
or
+ 2y + In Iyl
2
we obtain
dx or sin y dy
t sin 2x + e
+ 2e3x = e.
dy = (4x
1  .
x+e
Ixl + e or y = elx4
dy = x21n x dx we obtain y;
+ 3)2
= x + e or y =
Iyl = x 2 + e or y = elex.
9.
= elef(x)
+ e.
~e3x
=
4 cos y
2y
+3 =
cos2x dx
= x33In
Ixl 
(eY
eY
+ 1) 2
dy 
+ 5 + e.
= 2x + sin 2x + Cl
+ 1)1
= ~(eX
y
x
(
2)1/2
(
2)1/2
1/2 dy =
1/2 dx we obtain
1+Y
= 1+x
(1 + y2)
(1 + x2)
15. From
S dS
16. Frorn Q 1
70
= k dr we obtain S = er.
dQ
= kdt
we obtain In
IQ 
701 = kt
29
+ e or Q 
70
+ e.
+ 1)2 + e.
14. FrOl11
+ e.
4x
sin 3x
2
.
2
1
2
12. From 2y dy =  cos3 3x dx =  tan 3x sec 3x dx we obtain y = '6 sec 3x
13. From
~x3
= qt.
+ e.
To
= eCef(x) which
= ~(x + 1)3 + e.
y1
Ig(y)1 = f(x) + e.
I = ef(x)+c
t cos 5x + e.
)2 dy = dx we obtain _
= x
5. From  dy
y
of the form In
ec we obtain g(y)
we obtain y
_e3x dx we obtain y
y1
+ 1)2 dx
an expression
Exercses
2.2
= (1P + 1 _1)P
t
1
18. From N dN = (tet+2  1) dt we obtain In INI = tet+2  et+2  t + c.
19. From y  23dy = x  1 dx or (1 __ 5_) dy = (1 __ 5_) dx we obtain
y+
x+4
y+3
x+4
y5Inly+31=x5Inlx+41+c
or
dy = (1 + _5_)
x3
Y +2Inlyll=x+5Inlx31+c
21. From x dx =
or
n
1  y2
(X+4)5
y+3
=C1exy.
dx we obtain
(y_1)2 _
xy
(x _ 3)5  C1e
1
1
eX
1
1
22. From 2 dy =
dx = ()2
dx we obtain  = tan1 eX + c or y = .
y
eX + ex
eX + 1
y
tan1 eX + c
23. From
21
The
= 
xl
= 
x+1
or y = x.
1  x dx = ( 2"
1  1) dx we obtain
. In Iyl =  l' In Ixl = c or xy = c1e 1/x. Using
25. From 1 dy = 2y
x
x
x
x
y(l) = 1 we find C1= el. The solution of the initialvalue problem is xy = e11/x.
1
26. From 1 _ 2y dy = dt we obtain ~ In 11 2yl = t + c or 1  2y = ee2t. Using y(O) = 5/2 we find
C1= 4. The solution of the initialvalue problem is 1  2y = _4e2t
27. Separating variables and integrating we obtain
dx
Vf=X2
dy
J1 y2
=0
and
30
sin1xsin1y=c.
or y = 2e2t + ~ .
Exercises 2.2
Setting x = O and y =
V3/2
we obtain e =
1r
problem is sinl x  sinl y = 1r/3. Solving for y and using a trigonometric identity we get
1
y = sin ( sin x
28. From
1r)
+ '3
= x cos
1r
V3 V1=X'2
'31r + Vr.:":
1  x2 sin '3 = '2 +
2
.
X
x
dy dx we obtain
1 + (2y)2
 1 + (x2)2
1
+e
 tanl 2y =  tanl x2
2
2
Using y(l)
= O we find
C}
or
tanl 2y
+ tanl
x2 = el.
tanl 2y
+ tanl
x2
= ~.
4
1 = 21 + ee
1ee'
1
ee
= 2 + 2ee,
and e=.
1 = 3ee,
3e
1 lxl
3
1 + le4xl
3_ e4xl
2;:3 + e4xl
21 
In Iy 
1
In Iy
21 
+ 21 + In el
In Iy
=x
+ 21 + In e = 4x
In I e(y  2)
y+2
I = 4x
y  2
4
e = e x
y+2
Solving for y we get y = 2(e+X)/(eX).
~3e + x
3e  e4x
3  xl
= 2 ::3 + e4xl
31
Exercises
2.2
Y  1
+ 1)
y
.!.x dx
dy =
1
1
ClX
Another solution is y = O.
If y(O) = O then y = O.
1
(e) If y(1/2) = 1/2 then y = 1 + 2x .
i we obtain
1
1
Thus,y==.
1 + ~x
31. Singular solutions of dy/dx
(eX + eX)dy/dx
= y2 is y
Cl (2)
1
2Cl
= 4,
and
Cl
2'
2
2 + 3x
= x~
= O.
+ 10) + cscy
= C we get
2x
x2
dy
+ 10  coty csey dx = O,
2x
x2
+ 10
_ eosy_l_dy
siny sinydx
'
or
2xsin2ydx
 (x2
+ 10) eosydy
= O.
(x2
2xsin2 y
+ 10) eosy
we see that singular solutions oecur when sin2 y = O, or y = k7r, where k is an integer.
32
Exercises 2.2
33.
y
1. 01
0.0040.002
0.0020.004
0.002 0.004
0.98
0.97
y1
=x + c
dy
y1
and
)2 = dx. Then
y
1.02
x+c1
x+c
= 
0.0040.002
x  101
.
xlOO
0.99
0.98
dy
)2
y  1
+ 0.01
= dx. Then
1
x+c
y=l+tan.
10
10
Setting x = O and y = 1 we obtain c = O. The solution is
10tan110(y1)=x+c
1.0004
and
1. 0002
1
x
y = 1 + 10 tan 10 .
0.9998
0.9996
dy
)2
y  1
 0.01
= dx. Then
1.0004
51n 10Y 1
111
10y 9
= x + c.
1.0002
111 = x.
0.9998
10y 9
0.9996
33
Exercises 2.2
37. Separating variables, we have
dy
Y  y3
dy
y(l _ y)(l
+ y) =
(1
1/2
y + 1_ y
1/2
 1 + )dy
dx.
Integrating, we get
In Iyl 
"2 In 11  yl  "2 In 11 + yl = x + c.
When y
1
1
lny  ln(y  1)  ln(y
2
2
Letting x
x
= Oand
+ 1) = In
y
~
Vy2  1
= x + c.
2ex /
v'4e2x
3, where
When O < y
< 1 we have
1
In y  ln(l
2
Letting
00
1
 y)  ln(l
2
+ y) = In
y
~
VI  y2
= x + c.
= eX/v'e2x + 3, where
When 1
Letting x = O and y
where 00 < x < oo.
When y
1
 y)  ln(l
2
= ~ we find c = In(l/V3).
+ y)
= In ~
y
VI _ y2
=x
+ c.
= ex/v'e2x+3,
< 1 we have
1
ln(y)  ln(l2
1
y)  ln(l2
y) = In
y
#=1 = x + c.
34
Y4(X)
2ex /
v'4e2x
 3,
Exercises 2.2
y
SX
2
2
2
2
2
4
4
4
SX
l/(y  3) _
(y  3)2 
dy/dx_
(y1)2
I
1
(y3)3'
> O or y > 3,
> 3.
2
y
8
(y  3) dy = dx
'21Y 2 y2 _ 6y
3y = x
+e
+ 9 = 2x + el
(y  3)2 = 2x
+ el
y = 3 ..j2x
2
+ el.
The initial condition dictates whether to use the plus or minus signo
When Yl(O) = 4 we have q = 1 and Yl(x) = 3 + ..j2x
When Y2(O)
+ l.
+ l.
= 4 we have el
= 1 we find e = 1, so y2
= (2x
= x2
+ l)dx.
+x
Integrating gives y2 = x2
 1 and y = vx2
35
+x
+ x + e.
When
Exetcises 2.2
y
7
4 3 
2"
1_1
2
3
4
5
+x (00, ~  h/5).
(e) Solving x2
O we get x
+ 2e3x
we get y
= ~ In 1(5 
2e3x).
y
1
2
(e) Solving 1(5  2e3x) = Owe get z = 1In(~), so the exact interval of definition is (00, ~ ln(~)).
41. (a) While Y2(X)
= V25
 x2 is defined at x
= 5
and x
xlv
= O, the
i= O the
graph is a square
(b) Since x/y is not defined when y = O, the initialvalue problem has no solution.
(e) Setting x
= vx2
+ 3,
where the positive square root is chosen because of the initial condition. The domain is all real
numbers since x2
Exercises 2.2
I+Y2 sin2 y)
= dx
y
3.5
+ y) = dx / ( .x + x).
4
2
To integrate
J dt / ( Vi + t)
8 x
we substitute
u2 = t and get
J u+u
du
~2
J l+u
_2_
du = 2ln 11+ ul
+ C = 2ln(1 + vX) + c.
+ vX) + C
or
+ vX) + ln cj.
= [C1(1+.x)  1J2.
(~~r
= 1.
=~
~ dy
=:}
v=
= dx
=:}
. 1
sm y
= x + C.
_y2
y = sin(c  x)
=
sin(x  c)
=
sin(x
+ Cl)'
= 1 are solutions.
(b) Por Ixl > 1 and Iyl > 1 the differential equation is dy / dx = ~
variables and integrating, we obtain
dy
dx
and coshl y = cosh1 x
vX2=l
37
/.JX2=l.
+ c.
Separating
Exercises 2.2
Setting x
explicit solution is y = x.
47. (a) Separating variables and integrating, we have
(3y2
+ l)dy =
(8x
+ 5)dx
and
y3
+ Y = 4x2
 5x
= y3
+ c.
+y+4x2
= 2 is f(0,2)
and to y(l)
= 10; to y(l)
= 4 is f(1,4)
3 is 31.
(2y
+ y2)dy
= (x  x2)dx
and
y
2131213
+ 3'y
= 2"x  3'x
+ c.
+ 2x3
= 67;
38
Exercises 2.3
dy/dx
+ 2x3
 3x2
= ~.
(x  x2)/(y2
Using implicit
2
 2y), which
4
2
= 2, we find x
2
4
6
+ 2x3
2y3  6y2
for x we get x
8
4
2
10
Exercises 2.3
+ 2y =
= e2x so that
+y
= ce2x for
dx
+ 4y
00
1 an integrating
factor is eJ 4 dx
00
x so that
d~ [e4Xy]
= 1x and y =
1 + ce4x
d~ [e2Xy] = O and y
4. For y'
for
3. For y'
00
d~ [e5Xy]
2. For y'
00
= e5x so that
Exercses 2.3
x3 an integrating factor is eJ 2xdx
+ 2xy =
6. For y'
for O < x
+ 11y =
7. For y'
00
< oo.
= eX so that
1
= x so that d [xy] = 1 and y = lnx
+ e
dx
8. For y' _ 2y
+ ~y = ~
for O < x
y = ~x3  ~x
12. For y'_
1 eX
= ?
2 z
x4 so that
ddx [x4y]
+~
x
y
x6
x4 and
< oo.
+ ce"
x+1
for O < x
for 1
= x2ex
factor is eJ[1+(2/x)]dx
so that
ce:"
x
2 .
16. For dx
dy
2
for O < y
=
< oo.
= y2 so that
eY + 2" + 2" for O < y < oo. The transient term is
Y
= 3x and y = ~ + ex2
[x2y]
+ (1 + ~) Y =
+ cy4
dx
= sinx and
dx
ce:"
2y6
eY
for O < x
2x + 3
and y = x  x+1
15. For ~~  ; x
.!!:_
+ 2x
sin 2x and y
+ ex4
so that
x
y
(1 + x)
+ l)e
x(x
= x2
dx
< oo.
+ ~y
40
2+e
2 .
:y [y4x]
4y and
Exercses 2.3
tan
x dx
= 2xex
dr
de
+r
ex + x+1
sec e
ce3x
y = e3x + x
for O < x
= (x+1)eX,
y] = sec2 x
.iy
x
1
(3 +~)
dx
and (x  l)y
< 7f/2.
.!!:_ [( sin x)
= 3(x + 2)1 + c(x + 2)4 for 2 < x < oo. The entire solution is transient.
x dx = sin x so that
cot
x+1
= sec e+tan
e so that ded
[r(sec e + tan
= 2 + cett2 for
00
e)] =
1 and
x1
x(x + 1) + c(x + 1) for 1
x+1
< x < l.
1
x
so that
2 e
.
x
.!!:_ [X dx
x+1
1y]  1
1
d []xy = e x and y = e
dx
x
+ e
x
= 2 and x = 2y2
+ cy
= _ex + 
.!!:_
dy
[~x]
y
sY'
i = ~ + ceRt/L
for
00
= eRt/ L so that
= io then e = io
!!_
dt
[ieRt/ L]
E eRt/ L and
L
28. For dT  kT = Tmk an integrating factor is eJ(k)dt = ekt so that dd [Tekt] = Tmkekt
dt
t
T = Tw. + cekt for 00 < t < oo. If T(O) = To then e = To  Tm and T = Tm + (To  Tm)t.
41
and
Exercises 2.3
2x
={
2le2x
+ ci ,
C2,
If y(O) = O then
C2
1/2
Cl
O:::; x < 3;
x> 3.
1"x
= ~ ~so that
O :::; x :::;3;
x> 3.
32. For y' + y =
x>l.
If y(O)
= 1 then
Cl
so that
ye
= 2 then
Cl
2e
x>l.
{~ex2
C2,
If y(O)
O :::; x :::;1;
1,
y= { 2elx1,
33. For y' + 2xy = f(x)
C2
+Cl,
O:::; x:::;
1;
x> l.
C2
1e + ~ so that
O :::; x :::;1;
x>l.
42
Exercises 2.3
~'
1+ x
 { x
1 + x2'
,2x
Y +y1 + x2
34. For
O:::;x:::;l;
1
x
( 1 +x
If y(O)
= O then
Cl
= O and
1
O:::; x:::;
2) y = { ~x2 + ci,
_~x2+C2,
x>l.
_
y 
C2 =
2 (1, + x2) ,
3
1
2 (1 + x
1;
2)  ,
x>
1 so that
,
l.
35. We need
P(x)dx
2X,
O:::;x:::;l
21nx,
x>
20
15
An integrating factor is
eJ
P(x)dx =
2x
e ,
10
O:::;x:::;l
{ 1/x2,
and
d [{ ye2x,
dx
y / x2 ,
O:::; x < 1
>1
Integrating we get
ye2x,
{ y / x2 ,
Using y(O)
= 3 we find
Cl =
+ Cl,
2x  1 + 4e2x, .
{
Y = 4x21nx + (1 + 4e2)x2,
C2
O:::; x :::;1 .
x > 1
= 2  1 + 4e2 = 1 + 4e2.
O :::; x :::;1
x >1
= 1 is ex2. Thus
d [_X2
e y 1 =e
_x2
dx
2
[X
eX y = Jo et
dt
erf(x)
and
2
Y = eX erf(x)
. 43
+ ce" .
+c
Then
Exercises
Frorn y(l)
2.3
= 1 we get 1= e erf(l)
Y
37. For y'
+ eXy =
+ ce,
1 an integrating factor is
d
When y'
= 1 we get e = e, so y
+ eXy
= ee
= eX
1
+ eX (erf(x)  erf(l)).
and
t
fox dt
ee y
t
= lox
o dt + c.
+ e 1e x .
dy
y
= _ex dx
In jyj
and
Thus
 lee y] = ee
dx
Frorn y(O)
+ eXy = e"
_ex
+ c.
= ele".
= 1 is a solution.
+ cx4
In this case,
= x5ex
 x4ex
+ cx4
+e
XQ
= e fxdxj(9x2)
= e1In(9x2)
)9 _ x2
and so
_:!_ [) 9  x2
dx
> 00
y]
=O
and
y = ";9 c
+y
=3
(y  3x
+ 5) = y + 3x
 2,
"
x2
we get
,
3c
3 c
3
y  _   _  y
e x4 x x3 x
44
Exercises
+ 3y = O. Now
2.3
= 6x3
+ x3
is xy'
+ 3y
y(l)
13
= 13  1/13
+ 2/13
= O. Since
= 3. In each
+ 3y = 6x3,
= O has
y(O)
solution y
= x3 for
00
3
(e) The first two initialvalue problems in part (b) are not unique. For example, setting y(2) =
23  1/23 = 63/8, we see that y(2) = 63/8 is also an initial condition leading to the solution
y = x3  1/x3.
44. Since eJ P(x)dx+c
= eJ P(x)dx
C1eJ P(x)dxy
C2
= cleJ
r=.
we would have
and
eJ P(x)dxy
C3
C1 = Xo
and so x
= xoeAlt.
cleAlt.
= XOA1e(A2Al)t
dt
C2
y=
= /\2
\ XOA1
\
 /\1
Alt
+ C2e A2t
45
Frorn
x(O) = Xo
Exercises 2.3
47. (a) Letting y = YI
+ Y2
+ Y2(XO)
+ O= a
=a
and
+ Y2
+ P(x)y
= (x),
y(xo) = a.
yl + P(x)y = h(x)
+ h(x),
(1)
y(xo)=a+,6
has a unique solution. Consider y = YI +Y2 Since y(xo) = YI(XO) +Y2(XO) = a+,6,
the initial condition of (1). Also
y satisfies
+ Y2)
so y satisfies the differential equation in (1). Thus y = Yl +Y2 is the unique solution of (1) and
y(xo) = a +,6.
Since
CIYI
yl + P(x)y
is a solution of yl
+ P(x)y
eX2 y
ft/2,
Y = eX 2
=_
_x2
(X
Jo
et2 dt = _ ft erf(x)
2
ft erf(x) + J1F)
2
CIa,
d [ _x2 1
dx e y =e
From y(O) =
y(xo)
(b)
= Clil(X),
+ c.
ft /2.
Thus
10sin x
3
x
46
Exercses 2.3
is x2. Thus
d [ 2 1
xy=ldx
Osinx
X
x 2 y = 10
y
From y(l)
la
sin t
dt
+C
10x2Si(x)
+ cx2.
Thus
(b)
x
1
2
3
4
5
(e) From the graph in part (b) we see that the absolute maximum occurs around x = 1.7. Using
the rootfinding capability of a CAS and solving y'(x) = O for x we see that the absolute
maximum is (1.688,1.742).
50. (a) Separating variables and integrating, we have
dy
. 2
Y
= sin x dx
Now, letting t
and
(X
2
In Iyl = Jo sin t dt
+ c.
;tiu we have
(X
Jo sin t2 dt = V
tt
so
y
cleJ;sint2dt = CIe#Jo.2!;x
Cl
sin(1ru2/2) du
= 5 and y
(b)
JO
5
JO
47
= cle#S(y2;;x).
= 5e#S(y2;;x).
Exercises 2.3
(e) Frorn the graph we see that as x
9.35672. Since limx_oo 5(x)
~, limx__:_ooy(x) = 5e.;;s
+ 00,
oo
S(x) =
~ 2.672.
(d) Frorn the graph in part (b) we see that the absolute maximum occurs around x
absolute minimum occurs around x = 1.8.
solving y'(x) = O for x, we see that the absolute mximum is (1.772,12.235) and the absolute
minimum is (1.772,2.044).
Exercises 2.4
1. Let M = 2x1
h'(y)
= 3y + 7, and h(y)
= 2x1
we obtain f
+ ~y2 + 7y
= x2x+h(y),
= c.
2. Let NI = 2x + y and N = x  6y. Then My = 1 and Nx = 1, so the equation is not exacto
h'(y)
= y,
and h(y) =
= 2y2x  3 and N
5. Let M
1=
4xy
Nx'
7. Let M = x2  y2 and N
= x2
exacto
8. Let M = 1 + lnx+y/x
1=
y + ylnx + h(y), h'(x) = 1 + lnx, and h(y) = x lnx. The solution is y + ylnx + xlnx = C.
9. Let M = y3  y2sinx
ix
11. Let M
10. Let NI
1=
so that My
From Ix = x3 + y3 we obtain
= ylny  exy and N = l/y + xlny so that My = 1 + lny + yexy and Nx = lny. The
48
Exercises 2.4
12. Let M
obtain
= 3x2y + eY and N =
f = x3y + xeY + h(y),
=
13. Let M
f
= xy
14. Let M
+ 2ex + h(y),
= 1
3/x
x so that My
= O,and
h' (y)
y and N = 1  3/y
x  31n Ixl + xy
f =
+ y + xy
+ h(y),
h(y)
+ x so that
h'(y)
= 1=
My
1  ~, and h(y)
From
Nx'
2xex we obtain
2xex
+ 2ex = c.
= 1
fx
+y
3/x
 31n Ixyl = c.
15. Let M
 2x3  2xex
we obtain
x
x3
1/ (1 + 9x2)
x2y3 
1/ (1 + 9x2)
we obtain
= c.
and N
= !x3y3
x3y2 so that My
!arctan(3x)
+ h(y),
3x2y2
Nx'
From
= x2y3 
fx
= O. The solution is
x3y3  arctan(3x)
so that
5y2x
My
= 2
+ ~y2
= tanx 
17. Let M
and N
sinxsiny
From
fx
+ cos x sin y
= In Isec xl
2xy+h(y),
= c.
so that
cosxcosy
2y we obtain f
My
sinxcosy
Nx'
From
= O. The
fx
solution
= c.
2y sin x cos x  y
= Nx. From fx
2ysinxcosx
+ 2y2exy2 and
 1+ 4xy eXY
My
 y
+ 2y2exy2 we obtain
2sinxcosx
N = x
+ 2exy2
f =
+ 4yeXY
ysin2 x  xy
= Nx'
+ 2exy2 + h(y),
h'(y)
= O, and
= c.
= 4t3Y15t2y
and N = t4+3y2t
so that My = 4t31 = Nt. From ft = 4t3Y15t2y
we
obtain f = t4y  5t3  ty + h(y), h' (y) = 3y2, and h(y) = y3. The solution is t4y  5t3  ty + y3 = c.
19. Let M
ft =
1/t2y/
l/t
(t2
+ 1/t2
+ y2) and
 y/ (t2
N = yeY +t/
= x2+2xy+y2
and N
we obtain f = !x3 + x2y +
!x3
(f) +
+ y2)2
h(y), h'(y)
= yeY,
21. Let M
+ x2y + xy2
+ x2y + xy2
+ y2) so that
In Itl ~  aretan
!x3
(t2
= 2xy+x21
xy2
h(y),
(f) +
so that
h'(y)
My
yeY
eY
= 2(x+y)
= c.
= Nx' From fx = x2+2xy+y2
= y. The general solution is
1.
49
Exercses 2.4
= eX+y
22. Let M
and N
= 2+x+yeY
= 1 then e = 3 and
so that My
= 2y + yeY
= 1 = Nx'
From Ix
= eX+y
we obtain I = eX+xy+
+ 2y + ue" + 2y + yeY 
eY = c.
eY
= 3.
23. Let M = 4y
24. Let M
3
1=4
+
h(y), h'(y) = 3' and h(y)
4y
y
t2
3
2
2y
Ni. From
It
= t/2y4 we obtain
3
. The general solution is 4 2
4y
2y
t2
= c.
If y(l)
=1
= ~
26. Let M
y2
+ ysinx
we obtain
= xy2  Y cos x
(1
+ h(y),
+ y2)
=
+ 4kxy3
and N
= 3y2 + 40xy3
1
2
l+y
1 then e
so that My = 2y
h'(y) =
From
Iy = 2x2y
+ 2xy2
cos x and N
cos x we obtain
I = x2y2
+ sin x = Nx'
From Ix =
'
we obtain k
= 10.
= 9/2.
+ h(y),
h' (y)
= O, and
h(y)
= O. The
= Nx'
solution of the
and N = (y2+2xyx2)/(y2+2xy+x2)
(x2+2xy_y2)/(x2+2xy+y2)
My = 4xy/(x
+ y)3 = Nx'
From Ix = (x2
+ 2xy + y2
so that
2y2
= x + 
x+y
c(x + y).
+ 3x2
+ h(y),
31. We note that (My  Nx)/ N = s.], so an integrating factor is eJ dx/x = x. Let M = 2xy2
and N = 2x2y so that My
h'(y)
= O, and h(y)
= 4xy =
Nx' Frorn Ix
= 2xy2 + 3x2
we obtain I = x2y2
+ x3
= c.
32. We note that (My  Nx)/N = 1, so an integrating factor is eJ dx = e", Let M = xyeX
50
+ x3
+ y2ex + yeX
Exercises 2.4
and N = xex + 2yeX so that My = xex + 2yeX + eX = N.
I = xyeX + y2ex + h(x),
xyeX
+ y2ex
= c.
Nx'
= 4y3, and h(y) = y4. The solution of the differentia1 equation is 3x2y3 + y4
cotx,
= 
we obtain I = 1n(sinx)+h(y),
h'(y)
equation is 1n(sin x) + y + ln y2
cscx.
Let M
= 1+2/y,
= c.
= 3, so an
= c.
= e3x.
Let
M
and
so that My
h' (y)
= O, and
= 3/y,
13 e3x
M = (y2
+ xy3)/y3
= l/y
 2ye3x + x
= 1/y3.
= c.
Let
+x
and
N = (5y2  xy + y3 sin y)/y3 = 5/y  x/y2 + sin y,
so that My = _1/y2
we obtain I
= x/y+~x2+h(y),
h'(y) = 5/y+siny,
and h(y) = 51n Iyl cos y. The solution of the differentia1 equation is x/y + ~x2 + 51n Iyl cos y = c.
37. We note that (My  Nx)/N
Let M = x/(4+x2)
and N = (x2y+4y)/(4+x2)
we obtain I = ~ 1n(4 + x2) + h(y), h'(y) = y, and h(y) = ~y2. The solution of the differentia1
equation is ~ 1n(4 + x2) + ~y2
= c.
38. We note that (My  Nx)/ N = 3/(1 + x), so an integrating factor is e3 J dX/(l+x) = 1/(1 +x)3.
M = (x2+y25)/(1+x)3
From Iy = y/(l
and N = (y+xy)/(1+x)3
+ x)2 we obtain I =
= y/(1+x)2,
h2/(1 + x)2
+ h(x),
so that My = 2y/(1+x)3
h'(x)
2
+ (1+x)2
2
+ (l+x)
51
+lnI1+xl
= Nx.
h(x) = 2/(1 + x)2 + 2/(1 + x) + ln 11+ z]. The solution of the differential equation is
y2
2(1+x)2
Let
= c.
Exercises 2.4
39. (a) Implicitly differentiating x3
3x2
+ 2x2
dy
dx
+ 2x2y + y2 = c and
+ 4xy + 2y
we obtain
dy = _ 3x2 + 4xy .
dx
2x2 + 2y
dy = O and
dx
= O and y = 2 in x3
(b) Setting x
setting x
= y = 1 we also
find c = 4. Thus, both initial conditions determine the same implicit solution.
(e) Solving x3
+ 2x2y + y2 = 4 for y
,.y
we get
and
.t(x, y)
=y
resulting in y dx
+ x dy = O. A solution
= (x/y)dy
= O. An integrating factor is
42.
y
yexy + y2  2
(b) Since fx
+ h'(x).
= M(x, y)
so that fy
Let M(x, y)
+x
(x2
= yl/2xl/2
= yl/2xl/2
y
yexy + y2  2'
+ y r1
we obtain f
= yl/2xl/2
Let N(x,y)
+ ~ (x2 +yr1.
d( /x2 + y2) =
x
/x2
Then x dx
+ y dy
= /
x2
+ y2 dx
x
+ y2
dx
+
/ x2
+ y2 = X + c is a
dx
y
'x2
+ y2
dy.
becomes
===
/ x2
+ y2
+ y2
+ y2
dy
= d( V/ x2 + y2 )
= dx.
44. To see that the statement is true, write the separable equaton as g(x)
M = g(x)
and N
= l/h(y),
+ c.
dx
+ dy/h(y).
Identifying
52
Exercises 2.5
Exercises 2.5
l. Letting y = ux we have
+ x(udx + xdu) = O
(x  ux)dx
+ xdu
dx
dx
x
 + du
In Ixl
=O
=O
+u =e
x In Ixl
+y
= ex.
2. Letting y = ux we have
(x
+ ux)
dx
+ x(udx + xdu) = O
+ 2u) dx + x du = O
(1
dx
du
+=0
x
1 + 2u
1
(1 + 2~) =
x2
x2
3. Letting x
+ 2xy
el
= el.
= vy we have
vy(v dy
+ y dv) + (y
 2vy) dy
+ (v2
+ 1)
vy dv
 2v
+ dy = O
vdv
(vIn Iv In
1.::11y
1)2
11 v 
dy
+ In Iyl = e
x/y  1
(x  y) In Ix 
53
+ In y
yl 
=e
y = e(x  y).
Exercses
4.. Letting x
2.5
vy we have
y(vdy
+ydv)
 2(vy
y dv  (v
+ y)
dy
+ 2)
dy = O
~_dy=O
v +2
=O
In Iv + 21 In Iyl = C
In
I~
+ 2ln Iyl = c
x
5. Letting
+ 2y
= cly2.
= ux we have
(U2X2 + ux2)
+ xdu) = O
dx  x2(udx
u2 dx  xdu = O
1
=c
u
x
ln ]r] +  = c
y
In Ixl
y In
6. Letting
y = ux
+
Ixl + x
= cy.
we have
(U2X2 + ux2)
dx
+ x2(u
(u2
dx
+ 2u)
dx
dx
+
x
1
In [z] + 2'ln
lul 
+ x du) = O
+ x du = O
du
=0
u(u+2)
1
2'ln
lu + 21 = c
x2u
=Cl
u+2
2Y = ci
X ;;
54
(y
;;
+2
Exercises 2.5
7. Letting y
= ux we have
+ x) (u dx + x du)
(ux  x) dx  (ux
+ 1)
(u2
+ x( u + 1) du = O
dx
dx
In Ixl + ~ In
=O
u + 1 du = O
u2 + 1
(u2 + 1) + tanl u = C
8. Letting y = ux we have
(x
+ 3ux)
dx  (3x
+ ux) (u dx + x du) = O
(u2
1) dx + x(u + 3) du = O
dx
x
u +3
(u1)(u+1)
In Ixl + 21n lu 
11 
du
=O
In lu + 11 = C
x(u  1)2
u+1
Cl
(y  x)2 = Cl(Y
+ x).
9. Letting y = ux we have
uxdx
+ (x + y'ux)(udx + xdu) = O
(x
+ xy'u)
(u3/2
du
+ u3/2
dx = O
+ ~) du + d:
=O
+ In Ixl =
2NY +
55
Exercises 2.5
10. Letting
y = ux
we have
(ux
+ Vx2 + u2x2
) dx  x(udx
+ x du) = O
xVI +'u2dx
 x2du = O
dx
du
x
In JxJ  In
VI + u2
lu + VI + u
u + VI + u
=0
=e
2 =
11. Letting
y = ux
we have
(X3  u3x3) dx
+ u2x3(udx + xdu)
=O
+ u2xdu = O
dx
dx
2
 +u du
In JxJ
12. Letting
= 2 we find el
y = ux
=O
+ "3u3 = e
3x31n JxJ
Using y(l)
elx
+ y3
elx3.
+ y3 = 8x3.
we have
(X2
+ 2u2x2)
dx  ux2(u dx
(1 + u2)
+ x du) = O
dx  ux du =
dx
udu
 1 +u2
=O
+ u2)
=e
In IxJ  ~ In (1
x2
1 + u2 = el
x4
= 1/2. The
= el
(y2
+ x2)
56
= y2 + x2.
Exercses 2.5
13. Letting y = ux we have
dx  z e" du = O
dx
ud u=O
e
x
In Ixl
eU = e
In Ixl  eY/x = e.
Using y(l)
= O we find
14. Letting x
= vy we have
y( v dy
+ y dv) + vy (1n vy
eY/x  l.
 In y  1) dy = O
+ v In u dy
y dv
=O
~+dy=O
vlnv
y
In 11nIvll
+ In Iyl = e
I~I
= el.
yln
Using y(l)
15. From y
= e we find el = e.
1
+ y1 = yx
x3w = x3
+ e or
and w
= y 3 we obtain.
yl we obtain dw
dx
eXw = _~e2x + e or yl = _~ex + ce:",
e3x
+y
= xy
+w
= _ex.
3
3
3
+ w
= .
An integrating factor is x so that
dw
dx
y3 = 1 + cx3.
17. From y I
dx
+ ee3x.
(1 + 2:.)
y = y2 and w = yl we obtain
x
+ eX + e or
1
1 2
19. From y  y = 2Y
and w
I
= e.
= y
y
= 1 +
dw
dx
(1 + 2:.)
w = l. An integrating
x
factor is
e
1 + _ex.
x
dw
we obtain dt
57
+ w
t
2'
Exercises 2.5
tw
t
Y
20. From y + 3 (
1+ t
2) Y =
et/y
= elX.
2t
4
dw
2t
 2t
.
2) Y and w = y3 we obtain   2 w = 2
. An mtegrating
3 1+ t
dt
1+ t
1+ t
(
1 then
e=
dx
2
2
e3x/2w = e3x/2 + e or y3/2 = 1 + ee3x/2. If y(O) = 4 then e = 7 and y3/2 = 1 + 7e3x/2.
du
23. Let u
24. Let u
or u2
du
dx
1u
or udu = dx. Thus !u2
u
= 2x +
el.
du
2
2
Then dx  1 = tan u or cos udu = dx. Thus
= 1 + dy/dx_
=x+e
+ Cl.
du
dx
. u )/(1  sin
in uu) we h ave
by (1  sin
tan u  sec u
=x
= 1 + dy/dx. Then   1 = sm u or
= dx. Multiplying
1  sin
2 ud u = dx or (2sec u  tan u sec u ) du = dx. Th us
cos u
+ C or tan( x + y)  sec(x + y)
1
.
du
1 + smu
=x
+ c.
du
= dy/dx  2. Then dx + 2 = 2 +
Fu or
1
yU du = dx. Thus
_eu
29. Let u
du
= dy/dx  1. Then dx + 1
1 + en or eudu
= dx. Thus
= x + C and _eyx+5 = x + e.
=x+
y so that duf d
du
1
1 + cos u
1  cos u
1  cos U
= 1 _ cos2 U = sin2 u
58
= ese
1 + cosu
2
u  ese u cot u
du = dx. Now
Exercises
so we have J(csc2 ucscucotu)du
x + c. Setting x
= O and
= J dx
= 7r / 4 we obtain
= v2 
= x+c.
= 3x+2y
so that duf dx
du
Then d
x
= 3+2dy/dx.
h 1.
= 3+
Now
u+ 2
5u + 6
1
5
=+so we have
J (~+
5
and
25u
2u
u +2
5u + 6
u +2
and 6 du
u +2
5u+
= 
+ 30
) du = dx
1
4
5(3x + 2y) + 25 ln 175x + 50y + 301 = x +
Setting x = 1 and y
= 1 we obtain e = t ln 95.
= dx.
25u
+ 30
1. The solution is
csc(x + y)  cot(x + y) = x +
30.Let u
Thus  cot(x+y)+csc(x+y)
2.5
c.
The solution is
144
= O as dy/dx = f(x,
y) where
M(x,y)
f(x, y)
=  N(x, y) .
The function f(x, y) must necessarily be homogeneous of degree O when M and N are homogeneous
of degree a. Since Mis homogeneous of degree a, M(tx, ty) = tO: M(x, y), and letting t
= l/x we
have
1
M(l, y/x) =  M(x, y)
xO:
or
M(x, y)
= xO:M(l,
y/x).
Thus
dy = f(x y) = _ xO:M(l,y/x)
dx
'
xO:N(l, y/x)
= _ M(l,y/x)
N(l,yjx)
= F (J!._).
x
~~ = G (~)
we use the fact that M(x, y) = yO:M(x/y,
the substitutions u
and G(x/y)
suggest, respectively,
Exercises
32. As x
2.5
+. 00,
Then, as x
e6x
+ 00,
+
O and y
e6x
+
+
2x
O and y
= Yl
+u
+ 3.
+
 ee6x) as (e6x
+ e)/(e6x
e).
2x  3.
and
dy
dx
dYl
dx
=+
du
dx
lead to
dYl
du
;;
+ dx
=P
+ Q(Yl + u) + R(Yl + u)
=P
or
du
dx (Q+2YlR)u=Ru.
This is a Bernoulli equation with n = 2 which can be reduced to the linear equation
dw
dx
+ (Q + 2YlR)w =
R
Q(x) = l/x,
+ (_2.x +~)
w = 1.
x
+ e or u = [~x + ex3
1.
An
Thus, y = ~ +u.
34. Write the differential equation in the form x(Y' / y) = In x + In y and let u = In y. Then du] dx = y' / y
= ln z
+u
which is
firstorder, linear. An integrating factor is e J dx] = l/x, so that (using integration by parts)
!!._ [2.u]
= ln
dx x
and
x2
~ =_
2. _ ln x
The solution is
In y = 1  ln x
+ ex
+ e.
ecx1
or
= 
Exercises 2.6
1. We identify (x, y) = 2x  3y
+ 1. Then,
_
for h = 0.1,
+ 0.7Yn + 0.1,
and
y(l.1) ~ Yl = 0.2(1)
y(1.2) ~ Y2 = 0.2(1.1)
60
Exercises 2.6
For h = 0.05
Yn+l = Yn
+ 0.05(2xn
 3Yn
and
y(1.05) ;:::::
Yl = 0.1(1)
y(1.1) ;:::::
Y2
y(1.15) ;:::::
Y3 = 0.1(1.1)
y(1.2) ;:::::
Y4
2. We identify f(x, y) = x
+ y2.
and
y(O.l) ;:::::
Yl
= 0.1(0) + O + 0.1(0)2 = O
y(0.2) ;:::::
Y2
For h = 0.05
Yn+l
and
y(0.05) ;:::::
Yl
= 0.D5(0) + O+ 0.05(0)2 = O
y(O.l) ;:::::
Y2 = 0.05(0.05)
y(0.15) ;:::::
Y3
y(0.2) ;:::::
Y4
+ O+ 0.05(0)2 = 0.0025
and
In Iy I
=x
+ c.
Thus y = Clex and, using y(O) = 1, we find c = 1, so y = e" is the solution of the initialvalue
problem.
61
Exercises 2.6
hO
 1
hO
 05
xn
Yn
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00
1.0000
1.1000
1.2100
1.3310
1.4641
1.6105
1.7716
1.9487
2.1436
2.3579
2.5937
True
Value
1.0000
1.1052
1.2214
1.3499
1.4918
1.6487
1.8221
2.0138
2.2255
2.4596
2.7183
Abs.
Error
% Rel.
0.0000
0.0052
0.0114
0.0189
0.0277
0.0382
0.0506
0.0650
0.0820
0.1017
0.1245
0.00
0.47
0.93
1.40
1.86
2.32
2.77
3.23
3.68
4.13
4.58
Error
xn
Yn
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
1.0000
1.0500
1.1025
1.1576
1.2155
1.2763
1.3401
1.4071
1.4775
1.5513
1.6289
1.7103
1.7959
1.8856
1.9799
2.0789
2.1829
2.2920
2.4066
2.5270
2.6533
True
Value
1.0000
1.0513
1.1052
1.1618
1.2214
1.2840
1.3499
1.4191
1.4918
1.5683
1.6487
1.7333
1.8221
1.9155
2.0138
2.1170
2.2255
2.3396
2.4596
2.5857
2.7183
Abs.
Error
% Re1.
0.0000
0.0013
0.0027
0.0042
0.0059
0.0077
0.0098
0.0120
0.0144
0.0170
0.0198
0.0229
0.0263
0.0299
0.0338
0.0381
0.0427
0.0476
0.0530
0.0588
0.0650
0.00
0.12
0.24
0.36
0.48
0.60
0.72
0.84
0.96
1.08
1. 20
1.32
1.44
1.56
1.68
1.80
1.92
2.04
2.15
2.27
2.39
Error
dy
y
Thus y = qeX
= 2xdx
and
In Iyl = x2
= 1, we find c = e1,
+ c.
2
problem.
hO
 1
h=O 05
xn
Yn
1.00
1.10
1.20
1.30
1.40
1.50
1.0000
1.2000
1.4640
1.8154
2.2874
2.9278
True
Value
1.0000
1.2337
1.5527
1.9937
2.6117
3.4903
Abs.
Error
% Rel.
Error
xn
Yn
0.0000
0.0337
0.0887
0.1784
0.3243
0.5625
0.00
2.73
5.71
8.95
12.42
16.12
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
1.0000
1.1000
1.2155
1.3492
1.5044
1.6849
1.8955
2.1419
2.4311
2.7714
3.1733
62
True
Va1ue
1.0000
1.1079
1.2337
1.3806
1.5527
1.7551
1.9937
2.2762
2.6117
3.0117
3.4903
Abs.
Error
% Rel.
0.0000
0.0079
0.0182
0.0314
0.0483
0.0702
0.0982
0.1343
0.1806
0.2403
0.3171
0.00
0.72
1.47
2.27
3.11
4.00
4.93
5.90
6.92
7.98
9.08
Error
Exercises 2.6
hO
 05
h=O.l
5.
h=O.l
Xn
Yn
xn
Yn
0.00
0.10
0.20
0.30
0.40
0.50
0.0000
0.1000
0.1905
0.2731
0.3492
0.4198
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.0000
0.0500
0.0976
0.1429
0.1863
0.2278
0.2676
0.3058
0.3427
0.3782
0.4124
h=O.l
h=O 05
Xn
Yn
Xn
Yn
0.00
0.10
0.20
0.30
0.40
0.50
0.5000
0.5250
0.5431
0.5548
0.5613
0.5639
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.5000
0.5125
0.5232
0.5322
0.5395
0.5452
0.5496
0.5527
0.5547
0.5559
0.5565
7.
h=O.l
9.
6.
Xn
Yn
Xn
Yn
1.0000
1.0000
1.0191
1.0588
1.1231
1.2194
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
1.0000
1.0000
1.0049
1.0147
1.0298
1.0506
1.0775
1.1115
1.1538
1.2057
1.2696
10.
I 1
Eul e r
~
10
1.0000
1.0500
1.1053
1.1668
1.2360
1.3144
1.4039
1.5070
1.6267
1.7670
1.9332
h=0.05
Xn
Yn
Xn
Yn
0.00
0.10
0.20
0.30
0.40
0.50
1.0000
1.1000
1.2159
1.3505
1.5072
1.6902
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
1.0000
1.0500
1.1039
1.1619
1.2245
1.2921
1.3651
1.4440
1.5293
1.6217
1.7219
hO
 05
Xn
Yn
Xn
Yn
0.00
0.10
0.20
0.30
0.40
0.50
0.5000
0.5250
0.5499
0.5747
0.5991
0.6231
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.5000
0.5125
0.5250
0.5375
0.5499
0.5623
0.5746
0.5868
0.5989
0.6109
0.6228
ARUngeKUlta
f\
\
,
!.
h;O.1
A RungC.Kjlt.
Yn
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
1\
Xn
1.0000
1.1000
1.2220
1.3753
1.5735
1.8371
h=O.l
1.00
1.10
1.20
1.30
1.40
1.50
. 1
Yn
0.00
0.10
0.20
0.30
0.40
0.50
h=O.l
8.
h=O 05
11.
h=0.05
Xn
Eulcr
\.
10
63
h=O.05
~RUngCKutta
~uler
5
Exercises 2.6
12.
h=O.25
h=O.1
h=O.05
Run g eKut t a
RungeKutta
Euler
Euler
Eul e r
2
2
13. Using separation of variables we find that the solution of the differential equation is y = 1/(1 x2),
which is undefined at x = 1, where the graph has a vertical asymptote.
h=O 1
Euler
h=O 05
Euler
hO
1
RK
h=O 05
RK
Xo
yo
Xo
yo
Xo
yo
Xo
Yo
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00
1.0000
1.0000
1. 0200
1. 0616
1.1292
1. 2313
1. 3829
1. 6123
1. 9763
2.6012
3.8191
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1. 00
1.0000
1.0000
1.0050
1. 0151
1. 0306
1.0518
1. 0795
1.1144
1.1579
1.2115
1.2776
1.3592
1.4608
1. 5888
1.7529
1. 9679
2.2584
2.6664
3.2708
4.2336
5.9363
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00
1.0000
1.0101
1.0417
1.0989
1.1905
1. 3333
1. 5625
1.9607
2.7771
5.2388
42.9931
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1. 00
1.0000
1.0025
1.0101
1.0230
1.0417
1.0667
1.0989
1.1396
1.1905
1.2539
1. 3333
1.4337
1. 5625
1.7316
1.9608
2.2857
2.7777
3.6034
5.2609
10.1973
84.0132
Because the actual solution of the differential equation becomes unbounded at x approaches 1, very
small changes in the inputs x will result in large changes in the corresponding outputs y. This can
be expected to have a serious effect on numerical procedures.
64
0.6
.._______
~.~~
01
"J4
.o
(b) For y
+ 2xy
= 1 an integrating factor is e
+ cex = 
2
d
2
2
x x = e" so that  [eX y 1 = eX and
dx
J2 d
erf (x)
+ cex. 2
erf (x).
we see that the solution has a relative minimum at
is a repeller for k
< O.
i dx
x
In y = 41n x
= O, so the
+ e = In x4 + e
65
5.
x even
ti
is even, Ois sernistable; when n is odd, Ois unstable. For daf d: = xn,when
7. ,._,~"",
"""
.",
_.....................
'
= 1.3.
"",.,_"
"",,.",
,,_
...
_... ..._
"\,,, "",,_.
,.
....."",
","
......
 ...1
(e) separable
(f) separable, linear in x, Bernoulli
(g) linear in x
(h) homogeneous
(i) Bernoulli
(j) homogeneous, exact, Bernoulli
(k) linear in x and y, exact, separable, homogeneous
66
cos x dx =
Y
2
y +1
dy
1 + 1sin 2x = In
1
x
2
4
2
===}
(2y
+ 1 + c
===}
(2
2x + sin 2x = 21n y + 1 + c.
10. Write the differential equation in the form y In; dx = (x In ;  y) dy. This is a homogeneous
equation, so let x = uy. Then dx = u dy + y du and the differential equation becomes
ylnu(udy+ydu)=(uylnuy)dy
or
ylnudu=dy.
===}
ulnlulu=lnlyl+c
===}
x(lnx lny)
===}
:::'lnl:::.I:::'=ln1Y1+c
y
y
y
. 1 equation
.
2 IS Bernou 11'
U'smg w
11. The diff
1 erentia
dy + y 2
y 3x2
1.
dx
6x + 1
6x + 1
dw
6
9x2
d + 6w
= .
An integrating factor is 6x + 1, so
x
x+1
6x + 1
d
2
[(6x + l)w] 9x
dx
 
===}
3x3
c
w +
  6x + 1 6x + 1
===}
(6x + 1)y3
y 3 we obtai
tam
== 3x3 + c.
Ix
= 3y2 + 2x we obtain
so h(y) =
3.
Iy
dt
= t,
so
1
dt [tQ] = t In t
===}
tQ =  25 t + 5t In t + c
===}
Q =  25 t + 5t In t + t'
67
2u
+1
u
+c
2u  ln 12u + 11 = 2x
+ cj ,
+ 2y + 2 
ln 14x + 2y
+ 31 = 2x + CI
2x
+ 2y + 2 
ln 14x + 2y
+ 31 = CI.
or
dy
15. Write the equation in the form d
x
8x
+ zrr:
x +4
2x
x +4
= 2.
An integrating factor is
x2
+ 4)4 ,so
= 2r2 cos e sin e + r cos e and N = 4r + sin e  2r cos2 e we see that M; = 4r cos e sin e +
= Ne so the differential equation is exacto From fe = 2r2 cos e sin e + r cos e we obtain
f = _r2cos2e + r sin ' + h(r). Then fr = 2rcos2e + sine + h'(r) = 4r + sine  2rcos2e and
h'(r) = 4r so h(r) = 2r2. The general solution is
16. Letting M
cos e
c/sinx.
7r < x
o.
= 2sin(71r/6)
= 77r/6.
68
= l. Thus, y = 1/sinx,
where
d; = _ 2 (t + 1) dt
y
1
 = (t
+ 1) 2 + e
The initial condition implies e = 9, so the solution of the initialvalue problem is
1
y
t2
+ 2t 
where
< t < 2.
 4
= dx
y'Y
=1=
+ 250
and
 xo, so that
 Xo
and
1
= 4(x
+ 2#0
 xo) .
+ c.
2y'Y = x
+ xy
dy
(0).
= ux.
=
Then
+ ux2(udx + x du) = O
dx
+ uxdu = O
dx
udu =x
1
'2u2 = ln jxj + e
=
y2
2" =  2ln j x j
+ q.
69
(b) Solving for y in part (a) and being sure that the initial condition is still satisfied, we have y = V2lxl(l
y
2
 In Ixl)1/2, where
e:::; x:::; e so that 11n [z] ;:::O. The graph ofthis function indicates that the derivative is not defined at x = Oand x = e. Thus,
2
< e.
21. The graph of Yl (x) is the portion of the closed black curve lying in the fourth quadrant. Its interval
of definition is approximately (0.7,4.3). The graph of Y2(X)
curve lying in the third quadrant. Its interval of definition is (00, O).
22. The first step of Euler's method gives y(1.1) ~ 9 + 0.1(1
more time gives y(1.2) ~ 9.4 + 0.1(1
+ l.lV9.4) ~
70
+ 3) = 9.4.
9.8373.
Equations
Exercises 3.1
1. Let
P = P(t) be the population at time t, and Po the initiaI population. From dP/dt = kP we
Using P(5) = 2Po we find k = ~ In 2 and P = Poe(ln2)t/5. Setting P(t) = 3Po
obtain P = pot.
we have
3 = e(ln2)t/5
==:::}
In3 = (1n2)t
5
==:::}
t = 5In3 ::::::
7.9 years.
In 2
=3
==:::}
In4 = (ln2)t
5
==:::}
t = 10 years.
in ProbIem 1 we obtain
10,000 = Poe(ln2)3/5
==:::}
Po = 10,000eO.61n2::::::
6597.5.
P = P(t) be the population at time t. From dP/dt = kt and P(O) = Po = 500 we obtain
P = 500t. Usng P(10) = 575 we find k = 1~In 1.15. Then P(30) = 500e31n1.15::::::
760 years ..
3. Let
N = N(t) be the number ofbacteria at time t and No the initiaI number. From dN/dt = kN we
obtain N = Not. Using N(3) = 400 and N(10) = 2000 we find 400 = Noe3k or = (400/No)1/3.
From N(10) = 2000 we then have
4. Let
==:::}
2000 = y,7/3
40010/3
o
==:::}
y, =
O
5. Let 1
6. From
N(t) be the amount of lead at time t. From d.Nf dt = kN and N(O) = 1 we obtan
ekt. Using N(3.3) = 1/2 we find k =
In 1/2. When 90% of the lead has decayed, 0.1 grams
7. Let N
71
Exercises
3. 1
= 0.1 we have
et(1/3.3)ln(1/2) = 0.1
==}
_t_In ~
3.3
= In 0.1
==}
In 1 2
8. Let N = N(t) be the amount at time t. From dN/dt = kt and N(O) = 100 we obtain N = 100t.
Using N(6) = 97 we find k = tlnO.97. Then N(24) = 100e(1/6)(ln097)24= 100(0.97)4;:::j88.5 mg.
9. Setting
N(t)
= 50
in Problem 8 we obtain
1
50 = 100e t
==}
kt = In 2'
In 1/2
==}
= (ln2)/k.
(b ) Since k = (ln2)/T
we have
12. From Example 3, the amount of carbon present at time t is A(t) = AoeO.00012378t.Letting t = 660
and solving for Ao we have A(660) = AoeO.0001237(660)
= 0.921553Ao. Thus, approximately 92%
of the original amount of C14 remained in the cloth as of 1988.
13. Assume that dT/dt
= k(T10)
so that T
10 + cekt. IfT(O)
= k(T5)
so that T
= 150 then
= 70
= 60
t = 3.06 minutes.
= 64.4611.
k(T  100) so that T = 100 + ce": If T(O) = 200 and T(l) = 220 then
c = 80 and k = In(39/40) so that T(t) = 90 implies t = 82.1 seconds. If T(t) = 98 then
t = 145.7seconds.
16. Using separation of variables to solve dT/dt = k(T Tm) we get T(t) = Tm +cekt. Using T(O) = 70
we find c = 70  Tm, so T(t) = Tm + (70  Tm)ekt. Using the given observations, we obtain
72
145.
Exercises
Then
/2
3. 1
ek
= (ek/2)2 = (110Tm)2
= 145Tm
70  Tm
70  Tm
(110  Tm)2
70  Tm
12100  220Tm
= 145 _ T
+ T'/'n=
10150  250Tm
+ T'/'n
Tm = 390.
The temperature in the oven is 390
0.
200  170et/50.
18. Frorn dA/dt
19. Frorn dA/dt
= 0 A/50 we obtain A
= cet/5D If A(O)
= 1000  1000et/100.
At t = 5, A(5) ;:::;
48.77 pounds.
dA
lOA
2A
= 10 = 10 we obtain A = 1000  10t
dt
500(105)t
100t
A(O) = O then c = 110' The tank is empty in 100 minutes.
20. Frorn
dA
21. From d =3
00
4A
2A
(
) =3weobtainA=50+t+c(50+t)2.
6 4 t
50 + t
t)2. If
IfA(0)=10then
+ c(100 
= 64.38 pounds.
of
3 gal/rnin and the solution is pumped out at arate of 2 gal/mn, the net change is an increase
of 1 gal/min. Thus, in 100 minutes the tank will contain its capacity of 400 gallons.
(b) The differential equation describing the amount of salt in the tank is A'(t)
=6
 2A/ (300 + t)
with solution
A(t) = 600 + 2t  (4.95 x 107)(300
+ t)2,
O < t :::;100,
as noted in the discussion following Example 5 in the texto Thus, the amount of salt in the
tank when it overflows is
A(100) = 800  (4.95
107)(400)2
= 490.625
lbs.
(c) When the tank is overflowing the amount of salt in the tank is governed by the differential
73
Exercses
3. 1
equation
dA
di
A(100) = 490.625.
Solving the equation we obtain A(t)
= 800
+ ee3tI400.
e = 654.947, so that
A(t) = 800  654.947e3tI400
When t = 150, A(150) = 587.37 lbs.
(d) As t
4
00, the
(400 gal)(2lbs/gal)=
(e)
800 lbs.
8oor~~~600
~200~40~0600t
+ Ri =
E(t), L
+ Ri = E(t),
,;=
'(0)' =
Smee
2
20
we obtai
tain e
E(t)
EoR
. t
EoLw
t+
RtIL
L2w2 + R2 smw  L2w2 + R2 cosw
ce
EoLw
= 20. + L2w2
+ R2
(l/e)q
= _200ee200t.
0.1472 amps. As t
4
= E(t), R
If i(O) =
00
we have q
4
10~0'
+ 2i =
If i(O) = Othen
+ 2i = Oand
74
= 60 and i
i = e2etl10.
= 60  60etllO.
Exercises
At t
= 20
we want C2e2
= 60  60e2 so that
i(t)=
C2 = 60
3. 1
< 20;
O ::s t
t>20.
60  60et/la
{ 60(e21)e~t/1O,
dq
dt
a  q /C = k~l + k2t
=}
C2
qI
C In I Ea  C
(Ea  qo/C)C
1/k2
k1
' so
(Ea  q/C)C
(kl + k2t)1/k2
+ k2t[ + Cl
k In [kl
2
q)C
( Ea  C
(
=
=}
qa)C
Ea  C
q
(
qa) ( kl )
Ea    Ea C C
k + k2t
q = EaC
29. (a) From mdv/dt
= mg
+ (qa 
kv we obtain v = gm/k+cekt/m.
EaC)
k)
( k + lk2t
kl
+ k2t
)1/k
1/ C k2
1/Ck2
and
t 00
vaT
gm)
kt/m
m.
+ c.
+ m ( va 
gm) .
k
+ 300.
31. When air resistance is proportional to velocity, the model for the velocity is m dv / dt = mg
kv (using the fact that the positive direction is upward.) Solving the differential equation using
separation of variables we obtain v(t) = mg/k
v(t) = _ mg
+ ce:"!":
From v(O)
+ (300 + mg)ekt/m.
k
75
= 300 we get
Exercises
3. 1
= 0.0025, m = 16/32
+ m (300 + mg)
k
(1 _ ekt/m).
s(t)
and
v(t) = 6,400
+ 6,700e0005t.
The maximum height is attained when v = O, that is, at ta = 9.162. The maximum height will be
s(9.162) = 1363.79 ft, which is less than the maximum height in part (a).
32. Assuming that air resistance is proportional to velocity and the positive direction is downward, the
model for the velocity is m dv / dt = mg  kv. Using separation of variables to solve this differential
equation we obtain v(t) = mg/k
+ cekt/m.
 ekt/m).
0.5, m
160/32
is Vo = 248.598. Her velocity with the parachute open (with time measured from the instant of
= 16 + 232.598e2t.
opening) is vp(t)
seconds after leaving the plane is five seconds after the parachute opens.
Twenty
time is vp(5) = 16.0106 ft/sec and she has fallen s(15) + sp(5) = 5514.02 + 79.9947 = 5594.01 ft.
Her terminal velocity is limt>oovp(t) = 16, so she has very nearly reached her terminal velocity
five seconds after the parachute opens. When the parachute opens, the distance to the ground is
15,000  5514.02
= 9485.98 ft.
Solving sp(t)
it
will take her approximately 9.88 minutes to reach the ground after her parachute has opened and
a total of (592.874 + 15)/60 = 10.13 minutes after she exits the planeo
33. (a) The differential equation is frstorder, linear.
eJ3bdt/(bt+ro) = (ro + bt)3. Then
!((ro
+ bt)3v]
= g(ro
+ bt)3
and
(ro
+ bt)3v
= :b (ro
+ bt)4 + c.
The solution of the differential equation is v( t) = (g/ 4b) (ro + bt) + c(ro + bt) 3. Using v(O) = O
we find c = gr~/4b,
so that
9
grt
gP(
k )
v(t) = 4b (ro + bt)  4b(ro + bt)3 = 4k ro + }
(b) Integratingdr/dt
= k/pweget
= kt/p+c.
Usingr(O)
76
gpr~
4k(ro + kt/ p)3 .
= ro wehavec = ro,
sor(t)
= kt/p+ro
Exercises
3. 1
= 0.007ft when t = lOs, then solving r(lO) = 0.007 for k/p, we obtain k/p = 0.0003 and
(e) Ifr
r(t) = 0.01  0.0003t. Solving r(t) = O we get t = 33.3, so the raindrop will have evaporated
completely at 33.3 seconds.
34. Separating variables we obtain
dP
p = k cos t di =>
If P(O)
= Po then
= k sin t + e ==> P =
= Po and P = po
el ek
sin
t.
sin t:
In IP I
7
00
as t
7
oo. If
7
oo.
7
36. The first equation can be solved by separation of variables. We obtain x = eleAlt.
we obtain el
= Xo
From x(O) = xo
e2
k(M  A)  k2A = Ofor A we find the equilibrium solution A = kM/(k1 + k2)' From
the phase portrait we see that lim,.....oo A(t) = k1M /(kl + k2).
~+~~A
Since k2
> O, the material will never be completely memorized and the larger k2 is, the less
77
Exercses 3. 1
!!_
dt
[e(k+k2)t
Al
=:}
=:}
= k M e(k +k2)t
1
e(k+k2)tA=
A  k k1Mk
1
UsingA(O)=Owefindc=A_,
+k2)A =
k1M
k1 + k2
2
e(k+k2)t+c
+ ce(k+k2)t .
k1M andA=
k1 + k2
kliV~ (1_e(k+k2)t).
k1 + 2
As t _,
00,
k1M .
k1 + k2
When x
> r/k, dx/dt < o. From the phase portrait we see that limt+oox(t) = r/k.
__~~~x
r
r 
39. It is necessary to know the air temperature from the time of death until the medical examiner
arrives. We will assume that the temperature of the air is a constant 65F. By Newton's law of
cooling we then have
dT
 65)
T(O) = 82.
dt
'
Using linearity or separation of variables we obtain T = 65+ct.
From T(O) = 82 we find c = 17, so
 = k(T
that T = 65 + 17ekt. To find k we need more information so we assume that the body temperature
at t = 2 hours was 75F. Then 75 = 65 + 17e2k and k = 0.2653 and
T(t)
= 65 + 17eO.2653t.
At the time of death, to, T(to) = 98.6F, so 98.6 = 65 + 17eO.2653t, which gives t = 2.568. Thus,
the murder took place about 2.568 hours prior to the discovery of the body.
78
Exercses
3.2
of both the room and the cream is 72F, and that the
of the coffee when it is first put on the table is 175F. If we let TJ(t) represent the
temperature
dt
k(TI
72),
which implies TI = 72 + CI t. At time t = O Mr. Jones adds cream to his coffee which immediately
reduces its temperature by an amount a, so that TI (O)
which implies
CI =
= TI (O) = 72 + CI,
fact that k
At t
< O we have
T2(5)  a
72 + 103e5k
< 72 + 103k
ak
41. Drop an object from a great height and measure its terminal velocity,
that
Vt
= gm/ k,
so k
Vt.
= gm/vi.
42. We saw in part (a) of Problem 30 that the ascent time is t = 9.375. To find when the cannonball
16t2
+ 300t = O, getting
Vi
= 18.75. Thus,
= v(18.75) = 300,
1,340,000 6,400t1,340,OOOe0.005t
Vi
When O < N
9.162.
= v(18.466) =
300.
Exercises 3.2
1. (a) Solving N(1  0.0005N)
t =
< 2000, dN/dt > O. From the phase portrait we see that 1imi<oo.N(t)
o
2000
79
= 2000.
Exercises 3.2
2000
dN
N(l  0.0005N)
= (~_
= dt
1
)dN
N  2000
1500
1000
and
500
In N ln(N
 2000)
= t + c.
= 2000+t
/(1
10
+ +t) =
2000et /(1
dN
di =
3. From
dP
di
= 1000 we have a
= P (101
107
p)
20t
15
= 10.
500a
500b)eat'
+ (a _
Sinee t~~ N =
50,000
= 1 + 9geO.7033t .
500
and P(O) = 5000 we obtain P = 0.0005 + 0.0995eo.lt so that
obtain
P
(t) = 3.929b
a/b
1 + (a/3.929b  l)eat
3.929a
 3.929b)eat
+ (a
c
1 + (c/3.929  l)eat .
At t = 60(1850) the population is 23.192 million, so
23 92 =
.1
or c = 23.192 + 23.192(c/3.929
1+ (c/3.929 _ 1)e60a
 1)e60a. At t = 120(1910)
c
91 972 = ,..,:=;,.
1 + (c/3.929  1)e120a
or c = 91.972 + 91.972(c/3.929
23.192)/23.192)2 (_c __
( (c c/3.929
 1
3.929
1)
= c  91.972
91.972
or
91.972(3.929)(c  23.192)2
The solution of this quadratic equation is c = 197.274. This in turn gives a = 0.0313. Therefore
P( )
t
197.274
= 1 + 49.21eo.0313t
80
Exercises 3.2
dX
150
8. From d = k(150  X)2, X(O) = O, and X(5) = 10 we obtain X = 150 k
where
t
150 t + 1
k = .000095238. Then X(20) = 33.3 grams and X ; 150 as t ; 00 so that the amount of A ; O
and the amount of B ; Oas t ; oo. If X (t) = 75 then t = 70 minutes.
9. (a) The initialvalue problem is dh/dt
h
10
= 8AhVh/Aw,
!!:!!._
Vh
= _ 8Ah dt
and
Aw
= (AwVH
Aw
2Vh = _ 8Ah t + c.
~~~~=~~500
4Aht)2/A;
4Aht)/Aw,
where AwVH
:s AwH/4Ah
(h) Identifying H = 10, Aw = 47l',and Ah = 7l'/576 we have h(t) = t2/331,776 Solving h(t)
= Owe
(fs72 /144)t
+ 10.
10. To obtain the solution of this differential equation we use h(t) from part (a) of Problem 11 in
Exercises 11.3 with Ah replaced by cAh. Then h(t) = (AwVH
4cAht)2/A;.
Solving h(t) = O
with c = 0.6 and the values from Problem 11 we see that the tank ernpties in 3035.79 seconds or
50.6 minutes.
11. (a) Separating variables and integrating we have
6h3/2dh
Using h(O)
20 we find c
= 5t
and
12h5/2
5
= 5t + c.
(800.;5  i~t)2/5. Solving h(t) = Owe see that the tank empties in 384.;5 seconds or
14.31 minutes.
(b) When the height of the water is h, the radius of the top of the water is r = h tan 30 = h] J3
and Aw
Ah r;;:;:.
7l'(2/12)2
Solving h(t)
= Owe
= (243t)2/5.
12. When the height of the water is h, the radius of the top of the water is g(20  h) and Aw =
47l'(20 h)2/25. The differential equation is
dh
dt
=_
Ah r;;:;:.
eAw V2gh
=_
7l'(2/12)2
~4h =
0.647l'(20_ h)2 /25 v 64f1
82
_~
Vh
6 (20 _ h)2 .
Exercises 3.2
(b)
Year
Census
Population
Predicted
Population
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
3.929
5.308
7.240
9.638
12.866
17.069
23.192
31.433
38.558
50.156
62.948
75.996
91.972
105.711
122.775
131.669
150.697
3.929
5.334
7.222
9.746
13.090
17.475
23.143
30.341
39.272
50.044
62.600
76.666
91.739
107.143
122.140
136.068
148.445
Error
Error
0.000
0.026
0.Q18
0.108
0.224
0.406
0.049
1.092
0.714
0.112
0.348
0.670
0.233
1.432
0.635
4.399
2.252
0.00
0.49
0.24
1.12
1.74
2.38
0.21
3.47
1.85
0.22
0.55
0.88
0.25
1.35
0.52
3.34
1.49
The modeI predicts a population of 159.0 million for 1960 and 167.8 million for 1970. The
census populations for these years were 179.3 and 203.3, respectively. The percentage errors
are 12.8 and 21.2, respectiveIy.
5. (a) The differential equation is dP/dt = P(lInP),
solution P
o.
6. From
then
dP
di
C
= P(abInP)
a
= b  In Ro .
7. Let X = X(t)
we obtain
be the amount of
1
b
In labInPI
e at
= t+Cl so that
P = ea ece
dt
w
. If P(O) = Po
If X(O) = O and
150  150e180kt
X(5) = 10 then X = 1 _ 2.5e180kt where k = .0001259, and X(20) = 29.3 grams. Now X ~ 60
as t ;
00,
81
Exercises 3.2
14. We solve
dv
dt
'
m = mgkv
v(O) = 300
= O we find
s(t) = ; In leos
fi
t  tanl 300{Jg
I +~
In (1
+ 90~~Ok)
Solving v(t) = O we see that ta = 6.60159. The maximum height is s(ta) = 823.84 ft.
15. (a) Let p be the weight density of the water and V the volume of the object. Archimedes' principle
states that the upward buoyant force has magnitude equal to the weight of the water displaced.
Taking the positive direction to be down, the differential equation is
dv
2
m dt = mg  kv 
pv.
Vk
Vk dv
(Vmg  pV)2  (Vkv)2
1
Vk ===~
Vmg  pV
tanh
1
Vkv
Vmg  pV
= dt
=t
+ c.
Thus
v ()t =
>
1 as t
> 00,
Jmg  pV
k
tan
h (vkmg
 kpV
m
t
J (mg 
+ y2 )dx + y dy
+ Cl )
pV) / k .
+ y2
and
N = y. Since M and N are both homogeneous of degree 1 we use the substitution y = ux. It
84
Exercises
3.2
'=""': dh
vh
5
dt
6
and
= 
r: 80
800y h  h
3
3/2
2 5/2
+ h
5
5
6
= t
+ e.
1024v5 seconds or 38.16 minutes. Thus, the tank empties more slowly when the base of the cone
is on the bottom.
13. (a) Separating variables we obtain
mdv
= dt
mg  kv2
~
dv
= dt
9 1 (kvjmg)2
y'ri0
Jkjmgdv
Vkg 1 (Vkvjy'ri0)2
0itanh1
V kg
=dt
Vkv = t + e
y'ri0
[g"g
tanh 1 Vkv
;;;:n;:; =
 t + el
ymg
v(t) =
Setting t
= O and
v = Vo we find el = tanh1(Vkvojy'ri0).
00,
83
Exercises
3.2
follows that
(x  Vx2
x
+ u2x2)
[(1  ,jI +
dx
2)
+ ux(udx + xdu)
+ U2]
dx
=O
+ x211.du=
udu
dx
1 + u2  ..j 1 + u2
udu
x
dx
..JI + u2 (1  ..JI + u2
(1  VI + u
2 ) = In x
+C
1
==:o:=CIX
1  ..JI + u2
r::;;
C2
1  VI t u~ = 
= 2C2X+ c~ = 4 (~)
(x
+ ~)
which is a famiIy of paraboIas symmetric with respect to the xaxis with vertex at (c2/2, O)
and focus at the origino
(b) Let u
x2 + y2 so that
Then
dy
1 du
y=x
dx
2 dx
and the differentiaI equation can be written in the form
1 du
   x
2 dx
= x+ yU or ~ du = yU.
2 dx
85
Exercises
3.2
2y'u = x
.jU=x+c
W3
= O and
V
W 42W
=dx
tanh1(2V42W)=x+c.
and
Using the facts that the hyperbolic tangent is an odd function and 1  tanh2 x
= sech2 x we
have
1
2"V4 
2W
= tanh( x  c) =  tanh(x + c)
2" W = tanh (x + c)
1
2"W
= 1  tanh2(x
+ c) = sech2(x + c).
= 102 for r2 we see that r2 = 20h  h2. Combining the rate of input of
with the rate of output due to evaporation, k11"r2 = k11"(20h  h2), we have dV/dt =
Using V = lO11"h2
111"h3,
= (20nh  11"h2)dh/dt.
Thus,
and
86
dh
dt
1  20kh
+ kh2
20h  h2
Exercises
and
3.2
h
10
8
+ 4000t 
chosen because h 2: O.
2000
71
too
too
(yt/ 2 + 4000t 
= 0.005 11m
too
19. (a)
t
u
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
P(t)
';.n~
5.308
7.240
9.638
12.866
17.069
23.192
31.433
38.558
50.156
62.948
75.996
91.972
105.711
122.775
131.669
150.697
179.300
t2 + 4000t  t2
t) = 0.005 lim ./!=.2;===:===::=::=:too v t + 4000t + t
4000t
tJ1
+ 4000/t + t
4000
= 0.005 lirn 
Q (t)
u.O.;:,
0.036
0.033
0.033
0.033
0.036
0.036
0.023
0.030
0.026
0.021
0.021
0.015
0.016
0.007
0.014
0.019
87
too 1
+1
= 0.005(2000)
= 10.
Exercses 3.2
(b) The regression line is Q = 0.0348391  0.000168222P.
Q
..
0.02
0.015
0.01
0.005
'~2 0:;''4
0:;::":6
0:;~80;:1'"70:1
7072:1:4::'""0
70
p
(e) The solution of the logistic equation is given in equation (5) in the texto Identifying a
0.0348391 and b
= 0.000168222 we have
P(t) =
aPo
.
bPo + (a  bpo)cat
P( ) _
0.136883
t  0.000660944 + 0.0341781eo.034839lt.
(e)
175
150
125
100
75
50
25
75
100
125
150
(f) We identify t = 180 with 1970, t = 190 with 1980, and t = 200 with 1990. The model predicts
P(180) = 188.661, P(190) = 193.735, and P(200) = 197.485. The actual population figures
for these years are 203.303, 226.542, and 248.765 millions. As t ; 00, P(t) ; a/b = 207.102.
20. (a) Using a CAS to solve P(l  P)
+ 0.3eP = Ofor P
we see that P
= 1.09216 is an equilibrium
solution.
(b) Since f(P)
= P(l
is increasing for Po
 P) + 0.3eP,
P(O)
= Po,
1
2
88
Exercises 3.2
p
(e) The curves for the second initialvalue problem are thicker. The
= 1. Comparing
1.5
1D t
dv
2
dt = mg  kv )
v(t) =
If
=O
v(O)
ffi
tanh
t.
yfk9)
;;:;t
To find the time of descent we solve s(t) = 823.84 and find td = 7.77882. The impact velocity is
Vt
= Jmg/k
dv = mg  2
mg
dt
Vt
m
or
dt
2)
Vt
1 v
= gt
+ ci.
Vt
= Oimplies
el
= O, so
gt
v(t) = Vt tanh  .
Vt
s(t) =
Vt tanh
gt dt =
V[
Vt
e2
= O, so
In (cosh gt )
+ e2.
Vt
s(t) = V[ In (cosh gt ) .
9
Vt
89
Exercises 3.2
for
Vt
gives
Vt ~
v2
gt
s(t) = _J_ln(cosh)
9
Vt
= 160, k
mg  kv. Using m
32. The integrating factor is eJ dt/640 = et/640 and the solution of the differential equation is
et/640v = J 32et/640dt
20,480  20,480et/640.
+ c. Since s(O)
= O,
the liquid we solve s(t) = 50075 = 425, obtaining ta = 5.16018. The velocity at the time of impact
with the liquid is
Va
m dv/dt = mgkv2.
= v(ta) = 164.482. When the object is in the liquid its velocity is modeled by
= (51,200v2)/1600.
Va
= __!!:!_
and
1600
J2 In 1 v  160J2
640
v + 160J2
1=
_l_t + c.
1600
eV2t/51.8443 or
_ v  160J2
v + 160J2
= 226.274,
= eV2t/51.8443
= O we
To find when the object hit s the bottom of the tank we solve s(t) = 75, obtaining tb = 0.466273.
The time from when the object is dropped from the helicopter to when it hits the bottom of the
tank is ta + tb = 5.62708 seconds.
90
Exercises 3.3
Exercises 3.3
1. The equation dxf dt.
dx]
A1dt
+ c from which
C1XO so that x =
xoeAlt.
= Xo
we find
.5._
dt
Y
= A1XO
eA2t
y]
[eA2t
e(A2AI)teA2t
A2  Al
Using y(O) = O we find c2 = A1XO/(A2
_
y,
A1xOe
_A t
1
so that
= A 1x Oe(A2AI)t + c2
+ C2eA2t
A1XO
A2  Al
eAlt
+ c2eA2t
 Al). Thus
A1XO (Alt
e
/\2  Al
e A2t) .
dt
_ eA2t)
A2  Al
Integrating we find
x,
y,
20
of substance A is so small that it contributes very little to substance B. Now we see from the graph that
y(50) ~ 16.2 and y(191)
~~2~5~5~0~7~51~0~012~515~0~
91
Exercses 3.3
4. Suppose that the series is described schematically by W ==> AlX ==> A2Y
Al, A2, and A3 are the decay constants for VV, X and Y, respectively, and Z is a stable elemento
Let w(t), x(t), y(t), and z(t) denote the amounts of substances W, X, Y, and Z, respectively. A
model for the radioactive series is
dw
= AlW
dt
dx
 = AlW  A2X
dt
dy
dt = A2X  A3Y
dz
dt = A3Y
5. The system is
11121
Xl = 2 . 3 + 50 X2  50 Xl . 4 =  25Xl + 50 X2 + 6
I
1
1
1
2
2
x2 = 50XI 4 50X2  50X2' 3 = 25xl  25x2.
11
x2 = Xl'
100
e, respectively,
so that
1
1
3
100Xl 6= 50X2  50Xl
1
6 + X3
100
1
 X2
100
1
. 2  X2
100
I
1
1
1
x3 = 100x2' 5  100x3  100x3 4=
. 5 = Xl
50
1
+ X3
100
 X2
100
1
1
20X2  20X3.
7. (a) A model is
dXl = 3 . X2 _ 2 . Xl
dt
100  t
100 + t '
Xl(O) = 100
dX2 = 2 . Xl _ 3 . X2
dt
100 + t
100  t '
X2(0)
= 50.
(b) Since the system is closed, no salt enters or leaves the system and Xl(t) +X2(t) = 100+50 = 150
for all time. Thus Xl = 150  X2 and the second equation in part (a) becomes
dX2
2(150  X2)
=
dt
100 + t
3X2
100  t
300
100 + t
=
2X2
100 + t
or
dX2
(2
3)
+ 100 + t + 100  t
which is linear in X2. An integrating factor is
di"
X2 =
300
100 + t '
e2ln(IOO+t)3ln(IOOt)
= (100 + t)2(100  t)3
92
3X2
100  t
Exercises 3.3
so
+ t)2(100
+ e] .
Thus
X2
1
2
1
]
= (100300
+ t)2 [c(100  t) 3  "2(100t) + "2(100 + t)(100  t)
300
3
1
(100 + t)2 [c(100  t) + t(lOO  t) .
At t
= 30, X2 = (300j1302)(703c
+ 30
8. A model is
dXI
di
dX2 =
di
1
(4 galjmin)(O lbjgal)  (4 galjmin) ( 200XI
lbjgal )
.
(4 gal/min)
dX3 = (4 galjmm)
.
di
(1200XI
(1150X2
)lbjgal ) lbjgal
. (1150X2 lbjgal )
. . ( 100X3
1 lbjgal )
(4 galjmm)
(4 galjmm)
or
dXI
1
= Xl
dt
50
dX2
1
= XIdt
50
dX3
2
 = X2dt
75

2
X2
75
1
X3
25
Over a long period of time we would expect Xl, X2, and X3 to approach Obecause the entering pure
water should flush the salt out of all three tanks.
9. From the graph we see that the populations are first egual
x,y
and y are
both 45.
5D
93
IDD
Exercises
3.3
10. (a) The population y(t) approaches 10,000, while the population
k,
l~k
1 Jo
10
20
; 1
1
10
1:}::1
II
5
11. (a)
E:
t..
I"'t
20
:1
10
20
10
20
1'" t
1'" t
(b)
1:
1 1
20
(e)
20
1 I"'t
40
20
40
(d)
1 1 1'"
40
In each case the population x(t) approaches 6,000, while the population y(t) approaches 8,000.
12. By Kirchoff's first law we have il = i2 + i3. By Kirchoff's second law, on each loop we have E(t) =
Li~ + R1i2 and E(t) =
u; + R2i3 + ~q
+ Li~ + Rli2 =
1
E(t)
+ R2i3 + Ci3 = O.
94
= q'
= CRli2  CR2i3
Exercses
13. By Kirchoff's first law we have il = i:
+ i3.
3.3
di2.
E(t) =
2lRl
E(t) =
2lRl
+ i; di + 22R2
and
.
di3.
+ L2di + 23R3.
di2
..
Lldi + (Rl + R2)22 + Rl23 = E
L2 di3 + Rli2 + (Rl + R3)i3 = E.
dt
14. By Kirchoff's first law we have il = i2
E(t) =
u; + Ri2
and E(t)
+ i3.
i3
= q'
+ i(t) + r(t)
persons, s(t), is proportional to the number of contacts between the number of people infected and
the number who are susceptible; that is, ds/dt
= klSi.
Next, the rate of change of the number of persons who have recovered is proportional to the number
infected; that is, dr / dt = k2i where k2 is positive since r is increasing. Finally, to obtain di/ dt we
use
d
d
(s+i+r)=n=O.
dt
dt
This gives
di
ds
.
.
dt
= dr
=
k22 + klSZ.
dt dt
The system of equations is then
ds
.
= klS2
dt
di =  k''22
+ k lS2.
dt
dr
 = k2i.
dt
A reasonable set of initial conditions is i(O) = io, the number of infected people at time O, s(O) =
n  io, and r(O) = O.
16. (a) If we know s(t) and i(t) then we can determine r(t) from s
95
+ i + r = n.
Exercses 3.3
(b) .In this case the system is
ds
= 0.2si
dt
di
O.7z+ 02'
z=:
. sz.
s(O) = 10  io since r(O) = O and i(t) + s(t) + r(t) = O for
all values of t. Now k2/kl = 0.7/0.2 = 3.5, so we consider initial conditions s(O) = 2, i(O) = 8;
s(O) = 3.4, i(O) = 6.6; s(O) = 7, i(O) = 3; and s(O) = 9, i(O) = 1.
io,
s.i
s.i
s.i
10
10
10
10
s.i
101
101
We see that an initial susceptible population greater than k2/ kl results in an epidemic in the
sense that the number of infected persons increases to a maximum before decreasing to
the other hand, when s(O)
<
k2/kl,
As t
+ 00
o.
On
+
C and y(t)
dx
= kl(y  x)
dt
dy
dt = k2(X  y),
96
+
C,
x,y
x(O)
Exercises 3.3
where kl
dt
dy
dt =kl(yx)k2(Xy)
dz
dt = kl(Z)
 k2Z
CIe(k+k2)t.
Now
and
= (xo  yo)k1
k1
y(t)
e(k+k2)t
+ k2
+ k2
+ XOk2 + yok1
k1 + k2
+ XOk2 + YOkl
kl + k2
This makes intuitive sense because the limiting concentration is seen to be a weighted average of
the two initial concentrations.
97
3.3
Exercises
xI,x2
30
20
10
50
100
1. From
dP
di
2. Let A
= A(t) be the volume of C02 at time t. From ddA = 1.2  ~ and A(O) = 16ft3 we obtain
= 0.018P
and P(O)
t
4
A = 4.8 + 11.2et/4. Since A(10) = 5.7ft3, the concentration is 0.017%. As t ......00 we have
A ......4.8 ft3 or 0.06%.
< 4, 6 :::;t < 10, and 12 :::;t < 16, no voltage is applied to the heart and E(t) = O. At
the other times the differential equation is dE/ dt =  E / RC. Separating variables, integrating,
and solving for E, we get E = ket/RC, subject to E(4) = E(10) = E(16) = 12. These initial
conditions yield, respectively, k = 12e4/RC, k = 12elO/RC, and k = 12e16/RC. Thus
O :::; t < 4, 6:::; t < 10, 12:::; t < 16
O,
E ( t ) 
12e(4t)/RC
12e(lOt)/ RC,'
12e(16t)/ RC,
(b)
4:::;t<6
10 :::;t < 12
16:::; t < 18.
10
\\\
4
10 12
16 18 '
98
+ (Co  Cs)ekAt/V,
dt = k[T 
T2  B(TI  T))
dT
(1 + B)T  (BTI
+ T2)1= kt + e
(b) Since k
e3
di
40  2i
___!!:!__.
10 
+ T2 .
1+ B
140  2il
BTI + T2
l+B
.
variables we obtain
el
= 2/JIO.
= 4 or i
=k
we obtain dx = ~
dx
= 2k
= q(lO 
tt
2,
OS t
t2:10
20,
= Owhen e =
BTI + T2)
l+B
'( ) _ { 4t It

If x
Then
[1 + (y')2]
and
t~~Ts=T2+BTIB
7. From y
= k dt. Then
BTI + T2 + TI  T2 (1+B)t.
l+B
l+B
1
In
2
O:::;t
+ T2)J.
TI  T2
B and so
1+
t+oo
(e) SineeTs=T2+B(TIT),
(1 
+ T2)
 (BTI
i=ecx:
6. We first solve
= k[(l + B)T
dy. If y
t).
10
20
tt
2,
= 20. Thus
< 10 .
= k sin2 e then
G ~
and
x = ke 
.
'2k sin
2e + c.
= O.
8. (a) From y =  x  1+ el e" we obtain y' = y + x so that the differential equation of the orthogonal
. dy =  1 or dx + x = y. A n integratmg
. f actor 1Se
. y , so
f am1'1y 1Sd
x
y+x
dy
~[eYxl
dy
= yeY
==:;.
eYx
= yeY + eY + c
99
==:;.
= y + 1 + ceY.
1
=y2'
= y'2.
 = x + e
y
y=.
x+e
9. From ~~
dy
dt
= k 2XY we obtam.
lnlyl=
k2 I l+Cle
klln
k ti +c
(k
or
y=C2 l+Cle
v"
!~~)
( + ( !~~)
(:o~
2~~l)
~~l)
100 gal
min
gal)
( 5 min
(Xl lb)
100 gal
100 gal
25
Xl ~ .
min
1
lb
20 Xl min .
(1~) (2~)+
min
100 gal
(2~) =
(4 gal)
min
100 gal
dt
dX2
dt
1
2
14 + 100x2  25xI
X
1
20
1
20
X2.
100
]_X2~.
20
min
equilibrium solutions P = 1 and P = 4. The phase portrait is shown below and solution curves
are shown in part (b).
>
4 and 1
< Po <
For
~ = P(5  P) 
= _(P2
5P
+ 4) =
dP
= dt
(P  4)(P  1)
( ~~)dP=dt
P4
P1
~ In 1P  41
3
P1
= =t + c
P 4
3t
P _ 1 = ci e
Setting t
= O and P
(e) To find when the population becomes extinct in the case O < Po < 1 we set P = O in
P  4
=e
P 1
Po  4
Po1
3t
from part (a) and solve for t. This gives the time of extinction
t
= _ ~ In 4(Po  1) .
3
Po  4
< O. Thus, if Po < ~, the population becomes extinct (otherwise there would be another
equilibrium solution.) Using separation of variables to solve the initialvalue problem we get
101
(b) Solving P(5  P)  7 = for P we obtain complex roots, so there are no equilibrium solutions.
< for all values of P, the population becomes extinct for any initial condition.
Since dP / dt
O5)
Solving P(t)
3. (a) Without
v'3]t
 2
P(r  r P] K).
With initial
P( )
rPo
t = rPo/K+(rrPo/K)ert
Po
50
100
50
100
Po
= ~K
Poi K + (1  Poi K)ert
2
or
2Po
= 1.
Po + (K  Po)ert
This gives
= _~ ( Po ) = __1_
(
70,000
) ~ 1 4 rs
r In K _ Po
0.08 In 400,000 _ 70,000
9. y .
PI =
for h, we get h
K  JK2  4Kh/r
2
= Po
= ha = 4620.
102
= 200,000.
50
4. (a) Letting 1959 be year O and 1973 year 14, we obtain the
100
400
graph shown.
300
200
100
10
12
14 t
10
12
14 t
400
300
.. ..
= 0.11.
400
300
200
= 500,
20
40
60
80
100 t
20
40
60
80
100 t
250
200
150
100
50
103
_
Thcn y(O) = CI + C2 = o, y'(O) = CI  C2 = 1 so
2. From y = clx
so that cI
= 3/5
+ gex.
= c +
3. From y = CIX+C2xInx
4. From y
 c2ex.
The solution is y
= 3x  4x
= CI = 3, y'(l)
= CI +C2 = 1 so
In x.
C2cos x + C3sin x we find y' = C2 sin x + C3cos x and yl/ = C2 cos X  C3sin x. Then
Y(7!") = CI  C2 = O, Y'(7!") = C3 = 2, yl/(7!") = C2 = 1 so that CI = 1, C2 = 1, and C3 = 2. The
solution is y = 1  cosx  2sinx.
5. From y = c + C2x2 we find y' = 2C2X. Then y(O) = CI = O, y'(O) = 2C2 . O = O and y'(O) = 1 is not
possible.
Since a2(x)
= x is O at x = O, Theorem
= x2 and y
= CI = O,
y'(O) = 2C2 . O
= O so
CI
=O
= 2x2.
= xosinwt
Thus
.
+ Xl smwt.
x ( t ) = Xo coswt
=x
+ C2Wcoswto
= Xl
and
C2 =
Thus
X ()t =
= xo(coswtcoswto
= Xo cosw(t
+ sinwtsinwto)
104
w
+ Xl (sinwtcoswto
w
sinwt
 coswtsinwto)
+ C2wcoswt.
Exercises 4. 1
9. Since a2(x) = x  2 and Xo
10. Since ao(x)
= tan
x and Xo
00
= Othe
=t: /2
(e2 1)
= 1 so that el = e/
< x < 2.
< x < 7r /2.
and e2 = e/
(e2 1).
= el
= 1,
y'(l)
= 2e2
= 6 so that el
y=1+3x2.
13. From y
= el eX cos x
= eleX( 
= el = 1, Y(7r)
= ele7T
= el
1 and e2
= O, Y(7r) = ele7T
= el . O+ e2 . 0+3
y = CIX2 
(el + 3)x4 + 3.
= 1 and e2 = l. The
17. Since (1/5)5 + (1) cos2 x + (1) sin2 x = Othe functions are linearly dependent.
18. Since (1) cos 2x + (1)1 + (2) cos2x = Othe functions are linearly dependent.
19. Since (4)x
=2+
x and h(x)
= 2 + [z]
105
3X
Exercises 4.1
1+ x
21. The functions are linearly independent since W (1 + x,
22. Since (1/2)eX
+ (1/2)eX
x2)
X,
x2
1 2x
23. The functions satisfy the differential equation and are linearly independent since
W (e3x,e4X)
for
00
i O
= 7ex
= cle3x + C2x.
24. The functions satisfy the differential equation and are linearly independent since
=2
W(cosh2x,sinh2x)
for
00
+ C2 sinh 2x.
y = Cl cosh 2x
25. The functions satisfy the differential equation and are linearly independent since
W (eX cos 2x, eX sin 2x)
for
00
= 2e2x i O
= clex cos 2x
26. The functions satisfy the differential equation and are linearly independent since
W (eX/2, xex/2)
for
00
= eX
i O
+ C2xex/2.
27. The functions satisfy the differential equation and are linearly independent since
W (x3,x4)
for O < x
= x6
i O
28. The functions satisfy the differential equation and are linearly independent since
W(cos(lnx),sin(lnx))
for O < x
= l/x
i O
= Cl cos(lnx) + C2 sin(lnx).
106
= 2 i O.
Exercises 4. 1
29. The functions satisfy the differential equation and are linearly independent since
= 9x6 =J. O
W (x,x2,x2Inx)
for O < x
+ C2X2 + C3x21n
x.
30. The functions satisfy the differential equation and are linearly independent since
00
Cl
COS x
equation, and Yp = x sin x+ (cos x) ln(cos x) is a particular solution of the nonhomogeneous equation.
equation, and Yp
e2x and Y2
x2e2x
+x
and Y2
= lsx2  ~x
x1
Also, Y~2 = 2x
+ 3 and
+ 5y PI =
12e2x  36e2x
+ 15e2x =
_ge2x .
Y~2 = 2, so
+ 3) + 5(x2 + 3x)
= 5x2
+ 3x
 16.
(b) By the, superposition principle for nonhomogeneous equations a particular solution of y"  6y' +
5y = 5x2
+ 3x
 16  ge2x is Yp
is
36.
(a) YPI = 5
(b) YP2 = 2x
(e) Yp = YPI + YP2 = 5  2x
+ C2.
107
Exercises
(b)
4. 1
Since D3x2
= o, x2,
= O. Since they
+ C2X + C3.
(e) Since D4x3 = O, x3, x2, x, and 1 are solutions of y(4) = O. Since they are linearly independent,
+ C2X + C3.
Since D3x3
= 3! = 6,
yp = x3 is a particular solution of ylll = 6. Thus, the general solution is y = C1x2 +C2X+C3 +x3.
(f) By part (e), the general solution ofy(4) = Ois Yc. = C1x3+C2X2+C3X+C4.
= cx3 + C2x2 +
+ x4.
38. By the superposition principle, if Y1 = eX and Y2 = ex are both solutions of a homogeneous linear
differential equation, then so are
1
2(Y1
+ Y2)
eX + ex
2
= coshx
1
2(Y1  Y2)
and
eX  ex
2
= sinhx.
y
3x
= O.
< O.
< O then
92 = x3
and W(Y1,Y2)
13:
x31
3x2
= 1 x32
3x
= o.
_x312
3x
= O.
= x2 is zero at x = O.
(e) The functions Y1 = x3 and Y2 = x2 are solutions of x2y"  4xy' + 6y = O. They are linearly
This does not violate Theorem 4.3 since a2(x)
independent since W (x3, x2) = x4
(d) The function y
=1=
Ofor
00
(e) Neither is the general solution since we form a general solution on an interval for which
a2 (x)
=1=
= eSex+2,
108
Exercises 4.2
41. Since OYI + OY2 +
42. The solutions are linearly dependent. Suppose n of the solutions are linearly independent (if not,
then the set YI. Y2, ... , Yn, Yn+l is linearly dependent.
Exercises 4.2
In Problems 18 we use reduction of order to find a second solution. In Problems 916 we use formula
(5) from the texto
1. Define y = u(x )e2x so
y'
Therefore u"
y"
and
y"  4y'
O and u
+ 4y = 4e2xu" = O.
Y2 = xe2x.
2. Define y
= u(x)xeX
so
y' = (1  x)eXu
+ xexu',
y" = xexu"
+ 2(1 
x)eXu'
 (2  x)eXu,
and
y"
If w
= u'
Therefore w
3. Define y
or
+~ w = Owhich
x
u"
+ ~ u' = O..
x
= u(x) cos4x so
y'
= 4u
sin 4x
y"
= u" cos 4x
and
y"
+ 16y =
(cos4x)u"
= O or u"  8(tan4x)u' = O.
 8(sin4x)u'
If w = u' we obtain the firstorder equation w'  8(tan 4x)w = O which has the integrating factor
eSJtan4xdx = cos24x. Now
d
dx [(cos2 4x)w]
Exercises 4.2
Therefore w = u' = C sec2 4x and u
4. Define y
= C! tan 4x.
= sin 4x.
= u(x) sin 3x so
y' = 3u cos 3x
y"
and
y"
If w
+ 9y
= (sin 3x)u"
factor
e6Jcot3xdx
d
.
dx[(sm23x)w]
Therefore w = u'
= c csc2
=O
gives
(sin2 3x)w
= c.
= cot 3x sin 3x
= cos 3x.
+ u' coshx,
and
y"  y = (coshx)u"
If w
e2Jta.nhxdx
= cosh/
+ 2(sinhx)u' = O or
equation w' + 2(tanhx)w
= csech2 x and u
6. Define y = u(x)X
+ 2(tanhx)u'
= O.
x. Now
d
dx [(cosh2 x)w]
Therefore w = u'
u"
Cl
= c.
so
and
y"  25y
If w
= X(u" + lOu')
w'
= O or
u" + lOu' = O.
+ lOw = Owhich
elOx. Now
..!!:.... [elOXw] = O gives
dx
and u = clelOx.
elOxw = c.
so
and
9y"  12y'
+ 4y = ge2x/3u" = O.
110
= e5x.
Exercises 4.2
Therefore u"
= O and u =
+ C2.
CIX
Taking q = 1 and
C2 =
Y2 = xe2x/3.
8. Define y = u(x )ex/3 so
and
+ y'
6y"
 y
=e
(6u"
X/3
X/6.
+ 5u')
w' +iw
=O
or
u"
+ ~u' = O.
Now
Therefore w
= u'
d
 [eSx/6w] = O gives x/6w = C.
dx
and u = cleSx/6. A second solution is Y2 = eSx/6ex/3
= ceSx/6
we have
Y2=X4
e J (7/x)dx
x8
dx=x4
~dx=x41nlxl.
e J(2/x) dx
x4
dx
x6dx
1
= "5x3.
= x3.
= L] we have
Y2
= x2
= lnx
e J dx/x
(In z )? dx = lnx
dx
x(lnx)2
= lnx
(1lnx ) =1.
A second solution is Y2 = 1.
12. Identifying P(x) = O we have
e JOdx
=xl/21nx
x(lnx)2
Y2=x1/21nx!
(1__ ) =_x 2.
1/
lnx
we have
. !
Y2 = xsm(1nx)
e J =dx]
.
2'
2
dx = xsm(1nx)
x sin (lnx)
= [xsin(lnx)] [cot(lnx)]
= xcos(1nx).
111
x
2(
x sin lnx
2'
dx
= ex/2.
J dx =
Exercses 4.2
14. Identifyng P(x) = 3/x
Y2 =
we have
X2
le
cos(lnx)
 J 3dx/x
4
x cos
2(1
nx
dx
= x 2 cos(lnx)
3
X2(
x cos In x
dx
Y2 = (x
+ 1)
= (z + 1)
e J2(1+x)dx/(12xx2)
(x + 1)2
dx
2x  x2
+ 1)2 dx = (x
+ 1)
l
l 1
(x
= (x + 1) [__
(x
+ 1)
l [2 +
(x
e1n(12xx2)
(x+ 1)2 dx
1) 2  1 dx
2_  x] = 2  x2  x.
x+l
A second solution is Y2 = x2 +
16. Identifying P(x) = 2x/
Y2 =
(1 
+ 2.
x2) we have
e J 2xdx/(1x2)dx
e1n(1x2)dx
1_11  x2
dx
= ~ In 1 + xI
2
1 x
x)l.
so
+ u'e2x,
y' = _2ue2x
and
y"  4y
If w = u' we obtain the first order equation w'  4w = O whch has the integrating factor e4
J dx =
e4x. Now
d
dx[e4Xwj
Therefore w
u' = cx and u
c1x.
=O
gives e4xw = c.
A second solution is Y2
= cle2x + C2e2x  
so
y' = u',
y"
= u"
and
y"
112
= O.
Exercises 4.2
If w = u' we obtan the first order equaton w' + w = Owhch has the ntegrating factor eJ dx = e",
Now
d
.
dx [eXw] = O gives eXw = c.
Therefore w
u'
cex and u
clex.
We see by
and
y"  3y'
+ 2y =
eXu"  eXu' = O
or
u"  u' = O.
Now
d
dx [eXw] = O
gives
exw = c.
= u' = ce" and u = ce". A second solution is Y2 = eXex = e2x. To find a particular
solution we try yp = Ae3x. Then y' = 3Ae3x, y" = 9Ae3x) and 9Ae3x  3 (3Ae3x) + 2Ae3x = 5e3x.
Therefore w
Thus A
= 5/2
and yp
= ~e3x.
so
and
y"  4y'
+ 3y
= eXu"  2exu'
=O
or
u"  2u'
= O.
If w = u' we obtain the first order equation w'  2w = Owhich has the integrating factor e2 J dx =
e2x. Now
d
dx [e2xw] = O gives e2xw = c.
113
Exercises 4.2
2L (a) For mI constant, let YI
= emx. Then
Yl
= mIemX and
y't
ay~
+ bm; + e = o.
Since a quadratic equation always has at least one real or complex root, the differential equation
must have a solution of the form YI
= emx.
+ yb, + ye = O,
a
= emX
J e
J
ebx/a
2mX dx
e(b/a+2mtlxdx
emxe(b/a+2m)
bja
+ 2mI
bja
+ 2mI
e(b/a+m)
.
dx = xemx.
= 2i (etX
=
e tX)
2(eX  eX)
+y =
x2u"
+ 2xu'
+ u(x)
and
 x2u'  xu
y" = xu"(x)
+ xu =
and YI(X)
= x is a solution of the
+ 2u'(x).
O. If we make the
u', the secondorder linear differential equation becomes x2w'  (x2  x)w
114
= 0,
Exercises 4.2
which is separable:
dw =
dx
(1 ~)w
x
dw
:;;
= ( 1In w
=x
1)
~ dx
+e
 In x
eX
Then u'
= el
x
el eX/ x and u = el J e" dx / x o To integrate eX/ x we use the series representation for e" o
= elX
1(
eX
; dx
 1+x
x
el
+ x
x2
2!
3!
(~
+ 1 + 2.x + 2.x
x
2!
3!
=elx ( lnx+x+2(2!)x
+
+ ooo) dx
o o
o)dX
1
+ 3(3!)x
4)
( x In x + x 21+31x
2(2!)
+ x
+000
+ ooo
3(3!)
+ lO)y' + 10y
= xe"  (x
+ 10)eX + IOe"
= O,
J P(x)dx
X
dx = e
Yl
x+lnx10
= eX
e2x
dx = eX
x=t
eJ
10
dx
X
dx = e
eJ(1+l0/x)dx
e
dx
2
X
x10ex dx
5,040x6
= 3,628,800  3,628,800x
 30,240x5
5,040x6
720x7
90x8
10x9
720x7
115
90x8
10x9
151,200x4
xl0)ex
 1,814,400x2  604,800x3

xlOo
151,200x4
Exercses 4.2
1101
Y2
lO!
L n! xn
is a solution.
n=O
Exercises 4.3
= Oand m = 1/4 so that y = Cl + C2ex/4.
2. From m2  36 = O we obtain m = 6 and m = 6 so that y = cle6x + C2e6x.
1. From 4m2
3.
+ m = O we obtain
From m2  m  6 = O we obtain m
+ 2 = O we obtain
m2 + 8m + 16 = O we obtain m = 4 and m = 4 so that
m2  10m + 25 = O we obtain m = 5 and m = 5 so that y
4. From m2  3m
5. From
6. From
= Cle4x + c2xe4x.
+ c2xx.
= clx
= clex/4
8. From m2
+ 4m
9. From m2
+ C2e(2v's)x.
+ 2m + 1 = O we obtain
3m
+ 4 = O we obtain
y
From m3  4m2  5m
m = 3/4
= e3x/4
+ C2sin x/2).
15.
+ C2e2x/3.
+ C2sin J2 x/3) .
V23i/4 so that
+ ex/2
+ 3m + 9 = O we obtain
(C2cos J3 x/2
m = 1, m
y = Clex
+ C3sin J3 x/2)
= 3, and m
+ C2e3x + C3xe3x.
116
= 3 so that
Exercises 4.3
18. From m3 + 3m2
4m  12 = O we obtain m
+ C2e2x + C3e3x.
y = cle2x
19. From m3 + m2
2 = O we obtain m = 1 and m
= clet + et(c2cost
u
20. From m3  m2
4 = Owe obtain m
= 1
= 2 and m
iso that
+ C3sint).
= Ce x + C2xe x + C3X2e x .
= Owe obtain m
= O, m
+ C4sin V3x/2)
= V3 i/2 so that
18 = Owe obtain m
= 3, m = 3,
and m = J2 i so that
2m3
10m2 + m + 5
= Owe obtain m
= 1, m = 1, m
= 1, and m = 1, and
m = 5 so that
U=Cle
28. From 2ms  7m4 + 12m3 + 8m2
x
+C2re
r
+C3e
+C4re
= Owe obtain m = O, m
+cse
= O, m
Sr
= 1/2, and m
= 2 2i so that
r
and y= 2cos4x
~sin4x.
117
=2
Exercises 4.3
31. From m2  4m  5
and y'(l)
y =
telx
O we obtain m
+ c2es =
2, then clel
1 and m
+ 5c2es =
O, clel
5, so that y
2, so Cl
+ c2x.
= e/3, C2 =
clex
If Y(l) = O
es /3, and
+ txs.
= O we obtain m = 1/2 and m
= 5 then
= 1, ~Cl
= 3/2 so that y
= 5, so ci =
= clex/2+c2e3x/2.
Ify(O)
=1
= _ex/2+ lie3x/2.
33. From m2+m+2 = Owe obtain m = 1/2/7
i/2 so that y = ex/2 (c cos /7 x/2 + C2sin /7 x/2).
If y(O) = O and y'(O) = O then Cl = O and C2= O so that y = O.
34. From m2  2m + 1 = O we obtain m = 1 and m = 1 so that y = cex + c2xex. If y(O) = 5 and
y'(O) = 10 then Cl = 5, Cl + C2= 10 so Cl = 5, C2= 5, and y = 5ex + 5xex.
35. From m3+12m2+36m
= Owe obtain m = O,m = 6, and m = 6 so that y = Cl+c2e6x+c3xe6x.
and y'(O)
If y(O) = O, y'(O)
c +C2
+~C2
6C2
+ C3= 1,
+ gxe6x.
m = 2, and m =
+ 2m2
5m  6 = Owe obtain m
y
If y(O)
= O, y'(O)
= O, and y"(O)
1,
3 so that
= 1 then
1
1
+ _e2x
+ _e3x.
15
10
39. Frorn m2
y = C2sin
2x.
+ 1 = O we obtain
+ C2sinx.
m = i so that y = Cl COSX
+ 2 = O we obtain m = 1 iso
ci = 1 and Y(7r) = e1l'COS7r = =e",
40. Frorn m2  2m
Y(7r)
1 then
that y
eX(cl cosx
+ C2sinx).
If y(O)
1 and
no solution.
41. The auxiliary equation is m2  3 = O which has roots V3
= cleV3x + C2eV3x.
118
Exercises
4.3
we find CI = i(3 + 5V3) and C2 = i(3  5V3) so y = i(3 + 5V3)ev'3x + i(3  5V3)ev'3x. For
y = CIcosh V3x + C2sinh V3x the initial conditions imply CI = 1, V3C2 = 5. Solving for CI and C2
42. The auxiliary equation is m2  1 = O which has roots 1 and 1. By (10) the general solution is
y
= qeX
+ c2ex.
= clex
+ c2ex the
and C2= e2/ (1+ e2) so y = eX/(1 + e2) + e2ex /(1 + e2). For y = CIcosh x + C2sinh x the boundary
conditions imply CI = 1, C2=  tanh 1, so y = cosh x  (tanh 1) sinh x.
43. The auxiliary equation should have two positive roots, so that the solution has the form y =
cl1X + c22X. Thus, the differential equation is (f).
44. The auxiliary equation should have one positive and one negative root, so that the solution has the
eax (CIcos bx
46. The auxiliary equation should have a repeated negative root, so that the solution has the form
= O where k
= Xo and x'(O)
+ C2sinh(8t/JI).
= 1, x(t)
= cosh(4tv's).
::::::
17.7989 ft/s.
and CI represent arbitrary constants, and each may take on any real value.
= m3 +6m2
15m
= O.
The differential equation is not unique since any constant multiple of the lefthand side of the
differential equation would lead to the auxiliary roots.
52. A third root must be m3 = 3  i and the auxiliary equation is
(m + 21) [m 
(3 + i)][m  (3  i)] = (m +
1 (m2
2)
119
11
 6x + 10) = m3  2m2
+ 7m + 5.
Exercises 4.3
T4e differential equation is
11
+ 6m2 + m 
34 by [m  (4
+ i)][m
Y2 =
 ( 4  i) 1 = m2
+ 8m + 17 gives m
 2. Therefore m3 = 2
is the third root of the auxiliary equation, and the general solution of the differential equation is
+
O as x
+ 00,
+y
+ (l/x)y' + y
Solutions of
xy" + y' + xy = Oare obtained using an ODE solver and are shown below with the indicated initial
conditions.
y
y
1
1
0.5
0.5
1
1
y(l)
= O,
y'(l)
=2
y(l)
= 2,
y'(l)
=O
+ 1 = (m 2 + 1) 2 
2m2 = (m2
+ 1  V2 m) (m 2 + 1 + V2 m)
= O.
differential equation is
y(x)
= eV2x/2
(CI cos
V; x +
C2 sin
V; x) +
+ bm + C = Ohas
eV2x/2
solutions m
(C3
cos
V;
+ C4 sin
V; x) .
= O.
120
Exercises 4.3
If b2  4c
= o, then
= eemx + C2temx.
= o, y(l) = O
is y = O.
If b2
4c
Now y(O)
< O then
= O implies
CI = O and
y
= c2ebx/2 sin 4c  b2 x.
O implies ~
= nm or
= CI and
= y( 7l'/2)
= C27l'
/2, so that c
= C2 =
this case also, the problem possesses only the trivial solution.
Case III
When A
= O yields
Vf.. = 2k for k
CI = O, so that y
When
Vf.. 7l'/2
= C2sin Vf.. x.
= CICOS Vf.. x
is an integer multiple of
tt ,
a nonzero integer, the problem wiJI have nontrivial solutions. Thus, for A = 4k2 the
Vf.. is not
trivial solution.
58. Applying integration by parts twice we have
J eax f(x) dx
 ~ [~eaxf'(x)
a a
 ~Jeax!"(x)dx]
a
a2
a2
dx
= ~ eax f(x) 
:2
!"(x))
121
:2
eax f'(x).
4.3
Exercises
eax (f(x)
:2
f"(x))
dx
eax f(x) dx
or
f(x)
 ~ f"(x)
a
= kf(x),
If k
< 1, k
+ a2(k
 l)f(x)
= O.
=1 O, the solution of the differential equation is a pair of exponential functions, in which
case the original integrand is an exponential function and does not require integration by parts for
its evaluation. Similarly, if k
ax
+ b.
In this case a
single application of integration by parts will suffice. Finally, if k > 1, the solution of the differential
equation is
f(x)
= Clcos av'k=!
+ C2sin a..Jk=l
x,
and we see that the technique will work for linear combinations of cos ax and sin ax.
59. Using a CAS to solve the auxiliary equation m3  6m2
m2
= 0.270534,
+ 2m + 1 we find ml
solution is
y = CleO.l1024lx+ eO.647826x
(C2
cos 0.857532x
1.74806, m2
 5.34m2
+ C3sin 0.857532x).
+ 6.33m
 2.03
= O we find
general solution is
y = Cle 1.74806x
= 1/2  /3i/2,
m3 = 1/2
+ V7i/2,
+ 2m2
and m4
 m + 2 = O we find ml = 1/2 + /3
= 1/2  V7i/2. The general solution is
122
i/2,
Exercises
4..4
= cle4x
CI + C2 + C3 = 2
1
4CI
+ "2C2+ C3+ C4 = 6
16cI
+ C2 + C3+ 2C4 = 3
64cI
_
116
_ 918
58
4 4x
116 x/2
y =  75 e
 Te
64.
From m4  3m3
y
+ 3m2
 m
= O we obtain m
918 x
58 x
25 e  5xe
= O, m = 1, m = 1, and m
If y(O) = O, y'(O)
so CI = 2, C2 = 2, C3 = 2, C4 = 1/2,
= O,
= 1 so that
and
Exercises 4.4
1. From m2
yp
+ 3m + 2 = O we find
+ C2e2x
+ C2e2x + 3.
y = cle~x
2. From 4m2
+ 9 = O we find mI = ~
and we assume
From m2  10m
+ 25
. 3
= ~, yp
= ~ and
+ c2xx
and we assume
123
+ 25B
= 3.
Exercises 4.4
4. From m2
Ax
3x
y=C1e
+C2e
2x
1
x
C1e3x
+ B.
Yp
18
5. From tm2
y = C1e2x
6. Frorn m2  8m
+ c2xe2x + x2
+ 2"'
 4x
2A  8B
+ 20C
13E = O
6D+
16A
=O
+ 20B
=O
13D = 26
20A = 100.
Then A = 5, B
i,
= 4, C =
y
D = 2, E
e2x(c1 cos 4x
iD eX and
 ~~) eX.
+B
Then A
= 4,
= 4, C = j,
 ~) e3x.
sin 2x. Substituting into the differential equation we obtain 19  8B = 1 and
8A  19B = O. Then A
lis,
y=cle
=
4~5 '
YP =
 41iscos 2x
425
124
4~5
2xsm 8.
425
Exercises 4.4
= clex + C2 and we assume Yp = Ax.
Substituting into the differential equation we obtain A = 3. Then A = 3, Yp = 3x and
y =
Cl e"
+ C2 + 3x.
+
10. From m2
2m
O we find
+ Ex + Cxe2x.
Yp = Ax2
mI
and
2
m2 =
O. Then Yc
C2
and we assume
+ 2E =
5,
+ ~xe2x, and
+ C2 + _x2 + 2x + _xe2x.
+ ~ = O we find mI = m2 = ~.
+ Ex2ex/2. Substituting into the differential
11. From m2  m
Yp = A
cle2x
Then Yc
equation we obtain ~A
12. From m2  16
Yp = Axx.
and
13. From
m2
= O we find
mI
+4
Yp = Ax cos 2x
+ Ex
4A = 3. Then A
= ~,
E = O, Yp
y
14. From m2
Yp = (Ax3
= 4 and m2 = 4.
Then Yc = clx
+ C2e4x
and we assume
Then Yc =
Cl COS2x
C2 sin
2x and we assume
= ixcos2x,
ci COS2x
and
we obtain
2E
+ 4F
6A+8E
=O
=O
12D = O
4C
+ 2E
8E+6D
=3
=O
12A = l.
125
Yc = Cl COS2x
+ C2 sin 2x
and we assume
Exercises 4.4
Then A
=
/2 ' B
0, C
= ~ , D = 0,
and
y = C cos 2x
15. From m2
(Ax2
+ 1=
+ Bx)
cos x
we find mi
= i and
+ (Cx2 + Dx)
m2
+ 2C =
O. Then A
4,
YP = lOx
14 3
+ 75x +
5x
= ce
1 4
+ C2
10x
we obtain 20A
equation
0, C
0, D
= ~, YP =
we find mi = 5 and m2
Substituting
= ~, B =
= 1,
6B  10G
1 4
we obtain 4C = 0,
equation
1 2
l.
2'x cosx+2xsmx.
Y=C1COSX+C2sinx16. From m2  5m =
Ax4+Bx3+Cx2+Dx.
2x.
sin x. Substituting
+ 116x2sin
cos 2x
14 3
53
= 25io',D = ~~~,
697
+ 75x + 250x
i:,
= 2, 12A15B
 625x.
17. Prom m2  2m + 5 = we find m = 1+ 2i and m2 = 1 2i. Then Yc = eX (Cl COS 2x + C2sin 2x) and
we assume YP = Axex cos 2x + Bxe" sin 2x. Substituting
into the differential equation we obtain
4B = 1 and 4A
= O. Then A
= 0, B =
i, YP = ixex
y = eX(c cos 2x
18. From m2  2m
+2
+ 2B
= 1 and  2A
we find mI = 1 + i and m2
+B
+ Be2x
= 1
sin x. Substituting
sin 2x.
i. Then Yc
equation
we obtain
y = eX(C cos x
7 2
+ C2sin. x) + Se
X cos x
l.
 Se 2x sin x.
= 0, 2A = 1, 3G + 4D = 3,
YP =
"21 cos
x 
9 COS 2 X
25
and 4G  3D
2
2
25 sin x, an d
Y = ce20.
From m2
+ 2m
assume YP = A
 24 =
+ C2xe
2'1 cos x
we find m = 6
+ (Bx2 + Cx)x.
Substituting
= O.
Then
A = ~, B = 0, C = 295' D =
9
 25 cos 2x
and m2
4.
12
+ 25 sin 2x.
Then
126
~,
Yc
equation
cle6x
+ C2x
we obtain 24A
and we
= 16,
Exercises 4.4
2E+10C
and
6x
io'
2 (120 x
+ C2e4x  3" 
C=
2
lgo' Yp = _~_(iox2
+ I~OX)x,
19)
4x
100 x e .
Yp = Ax2 + E cos x + C sin x. Substituting into the differential equation we obtain 12A
6E  C
and
= O. Then A = ~, E
1 2
6
1 .
x  37 cos x + 37 sin x.
"4
= cle2x+c2xe2x+c3e2x
= Cle2x +
3,
:f7,
= O, and
= CIe
+ C2xex + C3X2xe
= Owe find mI
= 3,
= 1, m2
C =
, Yp = x
 3  ~x3ex, and
23x e .
 x  3  3"x
t, C = ~, Yp = ~+ txex
+ ~xe2x, and
5 1
1
y = clex + C2e2x + C3e2x +  + xex + _xe2x
4
= 1, E
= 2, and 4A + C
= 1. Then A = 1, E
= 2, C
= 3, Yp = x2  2x  3, and
From m4  m2
= O we find mI
= m2 = O, m3 = 1, and m4
= 4, 2E
O, lOC  2D
= 1. Then
Yc
127
Exercises
4.4
27. We have Ye
we find A
=  ~.
we assume Yp
= A.
= j2 sin 2x  ~ .
equation we find A
=O
7, E
= 19, and
e=
37. Thus y
= cle2x
186 x/2
+5
e 
7x  19x  37.
29. We have Ye
= 200x/5
200  3x2
+ 30x.
Substituting
= cle2x + C2xe2x +
(ix3
into the
+ ~x2) e2x.
31. We have Ye
= Cl coshx
= Ax cosh x + Ex
= Atcoswt + Etsinwt.
= (Fo/2w2)sinwt  (Fo/2w)tcoswt.
128
Exercises 4.4
34. We have x = q cos wt
into the differential
+ C2sin wt
equation
X
+B
we find
A = Fo/(w2
= C1cos wt
B = O. Thus
 ,2) and
Fo
+ c2 sin wt + (2w ,
2) cos ,t.
35. We have Yc
differential
Cl
= (w 2 ,'
+ C2ex + C3xex
we find A
equation
2, B
and
+ Bx2ex + c.
Ax
Substituting
into the
C = ~. Thus
+ C3xe x + 2X + 9xex + 2x
equation
we find A
1
2
+ _x.
 12x2ex
12,
2) cosrz.
,
y = 11  llr.
Substituting
+ (2w
and we assume yp
y = C1 + C2ex
Frorn the initial conditions
2) coswt
and we assume
yp = Ax
t ' B =  ~, and
J3 x) + 4"x 
= ~.
+ B + Cxe2x.
Thus
8" + "3xe2X.
y=e 23
12
equation
we find A
1, B
and y(l)
= O we obtain
= O, and
1
5
+x+xe.
4
8
;:;)
+ C.
Substituting
2x
3
into the differential
+ C2sinx + x2  1.
From y(O)
=5
C1 1 = 5
(cos 1)C1 + sin(1)c2 = O.
Solving this system we find C1 = 6 and C2 = 6 cot 1. The solution of the boundaryvalue problem
is
we find A
+ C2sin x)
+ x2
and we assume yp = Ax
1 and B = O. Thus y
 1.
+ B.
Substituting
we obtain
C1 = O
129
From y(O)
= O and Y(7r) = 7r
Exercises 4.4
Solving this system we find Cl = O and C2 is any real number.
The solution
of the boundaryvalus
problem is
39. We have Yc
Cl COS2x
+ C2sin 2x
and we assume Yp
we find A
equation
O and B =
we have y = C3cos 2x
[O,7r/2]. On (7r/2,00)
=A
cos x
1 Thus
+ C4sin 2x.
+B
+ C2sin 2x + 1sin x
y = q cos 2x
on
q =1
1
we find Cl
1 and
1= C3
Hence C3 =
. Continuity
i = 2C4
differential
have y
= eX(q
cos3x
={
cos 3x
7r
+ c2sin3x)
+ C4sin3x).
is
O:::; x :::;tt /2
x> 7r/2.
= 2,
+ 3q
~.
Thus y
+ C2sin 3x) + 2 on
=
into the
[O,ttJ. On (7r,00)' we
O we obtain
= O.
= eX(
2 cos3x
+ ~sin3x) + 2 on
[O,7r].
= 7r implies
or C3 = 2e1l'(1
+ e"}.
continuity
of y' at 7r implies
problem
we find q = 2 and C2 =
of y at x
+ 2e1l' = c3e1l'
initialvalue
on [O,7r/2J.
+ ~sin 2x + 1sin x,
~ cos 2x + ~sin2x,
e1l'(2cos37r
or 2
cos 2x
Now, continuity
+ 1sin x
equation
= eX(c3
2x
y(x)
40. We have Yc
+ ~sin
= cos 2x
of y at x = 7r/2 implies
COS7r +
or 1 +
'3 + 2C2 = 2.
C2 = ~. Thus y
+ e")
we have C4 = ~e1l'(1
+ e"}.
is
y(x) =
ex ( 2 cos 3x
{ (1
+ sin 3x),
+ e1l')eX1l' (2
cos 3x
130
+ ~ sin 3x),
z > x,
Exercises
we find Y~ = Akx
and Y~ = Ak2x.
Substituting
4.4
equation we get
aAk2ekx
so (ak2
+ bk + c)A =
+ bAkx + cAx
= (ak2
+ bk + c)A
+ bm + e = O, A
we find Y~ = Akxx
= ekx,
= lj(ak2
+ bk + e).
Substituting into
+ (2ak + b)Aekx
+ (2ak + b)A
= (O)AxX
= (2ak + b)AX = x
and Y~ = Ak2x2x
and 2ak
+ 4Akxx
+ b = O.
= 2Ax.
ekx.
Since sin x is a solution of the related homogeneous differential equation we look for a particular
solution of the form YP = Ax sin x
+ Ex
cos x
+ e sin3 x.
we obtain
2a cos x
+ (6c 
YP = x cos x
43. (a)
(b ) f(t)
00
00
as t
as t
1 . 3
+ 4
sm x.
* 00
* 00
and Yp * O as t
and Yp *
131
00
as t
*
*
oo.
oo.
Exercises 4.4
(e) f(t)
(d) f(t)
+ (lOx2 
X 
1)e2Xsin2x.
12D = 2,
6A+SE
12A = 10,
SB+6D=1,
i,
2B
= 3,
+ 4F
= 0,
4C+2E=1,
= ~, E =
i, and
F =
k.
Thus, a particular
Cl
cos X
+ C2 sin x + C3X
COS X
+ C4X
solution of the form Yp = Ax2 cos X + Bx3 sin x. Substituting into the differential equation and using
a CAS to simplify yields
(Sa
This implies Sa
Exercises 4.5
1.
= (2D + l)(D
 2)y
+ ~x3 sin x.
=1
3x
Exercises 4.5
+ 8D)y
9. (D4
+ 2)(D2
= D(D
+ 16)y =
(D  2)2(D
+ 4)y
=4
+ 2)2y =
(x3  2x)X
= 8Dex/2
+ 64)(2cos8x
14. (D2
 2D
= 4ex/2
 4ex/2
 4ex/2 = O
= (D2)(2e2x15esx+5e2x+15eSX)
 5sin8x) = D(16sin8x
= 128 cos 8x
 40 cos8x)
= (D2)7e2x
+ 64(2cos8x
= 14e2x14e2x
 5sin8x)
 320 sin 8x = O
15. D4 because of x3
16. DS because of x4
19. D2
+ 4 because of cos 2x
21. D3(D2
22. D2(D2
23. (D
20. D(D2
+ l)(D
and cos 5x
+ 5)
+ 2D + 2) (D2
26. (D2
 4D
+ 5) because
30. D2  9D  36 = (D  12)(D
+ 3);
e12x, e3x
32. D2  6D
33. D3 
e3x sinx
D(D2
9)y
= O.
Then
y = ..._,___..,
ce3x + C2e 3x +
Yc
133
C3
=O
Exercises 4.5
and YP = A. Substituting Yp into the differential equation yields 9A = 54 or A = 6. The general
solution is
and YP
29 or A
29/5.
The
general solution is
D(D2 + D)y
= D2(D + l)y = O.
Then
Yc
and yp = Ax. Substituting YP into the differential equation yields A = 3. The general solution is
y
= Cl + c2e3x + 3x.
Then
Yc
Then
Yc
134
Exercises 4.5
and Yp = Ax + B. Substituting Yp into the differential equation yields 4Ax
Equating coefficients gives
4A
+ (4A + 4B)
= 2x
+ 6.
=2
4A +4B = 6.
Then A = 1/2, B = 1, and the general solution is
Then
Yc
and Yp = Ax2
+ Bx.
+ (2A + 3B) = 4x 
= 5.
3x
+ 'Jx
19
2

gx.
Then
Y
and Yp = Ax4
+ Bx3 + Cx2.
+ 2C
135
= O.
= 8x2.
5.
Exercises 4.5
Then A = 2/3, B
 ~x3
3
+ 8x2.
Then
!jc
and YP = Ax3
+ Bx2 + Cx + D.
Ax3
+ (B
 6A)x2
+ (6A
 4B
+ C)x + (2B
 2C
+ D) = x3 + 4x.
= 1, B
+D
=O
=4
= O.
+ 3)y
= O.
Then
!jc
and YP = Axe4x.
Substituting
+ 2D + 2)y
Then
!jc
136
= O.
=:
x.
Equating
Exercises 4.5
and Yp = Ax.
Substituting
coefficients gives A
= 5x.
Equating
(Cl COS X
O.
Then
Yc
and yp = Aex
+ B.
= 1
y
46. Applying D2(D
+ 2) to
and B
is
D2(D
+ 2)(D2 + 6D + 8)y
+ 2)2(D + 4)y
= D2(D
= O.
Then
Yc
and YP = Axe2x
+ Bx + C.
2Ae2x
= 3e2x
+ 2x.
6B+8C=
Then A = 3/2, B = 1/4, C
O.
47. Applying D2
+ 1 to
+ x1  .3
4
16
+ 1)(D2 + 25)y =
O.
Then
y
137
 3B = 4ex  9.
Exercises
4.5
+ 1) to
COS 5x
+ C2 sin 5x + 4 sin x.
COS 2x
+ B sin x + C.
and Yp = A ~os x
3A cos x
= 4/3,
= 2.
B = 1, and C
 8.
 2.
Then
Yc
and Yp = Axx
+ Bx.
+ (14A + 49B)e4x
= xx.
+ 49B
= O.
= cle3x + C2xe3x
xx
49
138
+ _e4x
343
Exercises
Then
Yc
and Yp = AxeX
+ Be + Cx + D.
+ (5A
6Axex
 6B)eX
10Cx
+ (3C
6A
=1
 10D)
= xe" + x.
5A  6B
=1
10C
3C  lOD = O.
Then A
+C2e
Sx
1
"6xe
5
36e
1
3
10x100'
= D(D
 1)4(D
+ l)y = O.
Then
y = cl e x
'v'
Yc
and yp = Ax3ex
+ Bx2ex + Cxe" + D.
6Ax2
+ (6A + 4B)xe
+ (2B + 2C)e
+ 5.
D = x2ex
=O
+ 2C = O
D= 5.
Then A = 1/6, B
52. Applying (D
+ 1)3 to the
1 3 x
+ C2ex + "6x
e 
1 2 x
4x e
1 x
+ 4xe
+ 1)3(D2 + 2D + l)y
139
= (D
+ l)Sy
= O.
4.5
Exercises
4.5
Then
Yc
+ Bx3ex + Cx2ex.
and Yp = Ax4ex
12Ax2ex
Equating coefficients gives A
1~'
B = O, and C
+ 2 to
2D
general solution is
1
+ c~xex
+ _x
4ex
~
12
Y = c 1ex
53. Applying D2
= O. The
2D
+ 2)(D2
+ 5)y =
2D
O.
Then
y
+ Be"
+ 3Bex
sin x = eX sin x.
+ 10 to
2D
(D2
2D
sin x.
+ 10)
(D2
+D +
y = (D2
+ 10)
2D
(D
~f
y
= O.
Then
Yc
+ Be"
+ 27B /4)eX
sin 3x
9A  27 B
4
1.
= c ex/2 + c xex/2
1
4
_ex
225
140
28
cos 3x  _ex
225
sin 3x.
Exercises 4.5
55. Applying D2
+ 25 to
Then
y = Cl cos 5x
+ C2 sin 5x + C3X
cos 5x
+ C4X
COS 5x
Yc
and Yp = Ax cos 5x
+ Bx
y
56. Applying D2
+ 1 to
Cl COS 5x
+ C2 sin 5x 
2x cos 5x.
Then
y = Cl
COS X
+ C2 sin x + C3X
cos x
+ C4X
COS X
Yc
and Yp = Ax cos x
+ Bx
+ 1)2 to
COS X
. x + '2x
1 cos x
+ C2 sin
 2'
x sm x.
+ 1)2(D2 + D + 1)
= O.
Then
y = e.x/2
[Cl
/3 x
cos 2
cos x
.x
+ C6X sin
Yc
and Yp = A cos x+B sin zlCz cos x+ Dx sinx. Substituting YPinto the differential equation yields
(B
+ C + 2D) cos x + Dx
cos x
+ (A 
2C
+ D) sin x 
=O
B+C+2D
D=O
A 2C+D
=O
C = 1.
141
Cx sin x = x sin x.
Exercises 4.5
Then A
[Cl
=!(1 +cos2x)
/3
cos 
. /3] + ')~
+ C2 sm 
+ sm x
cos x
 x cos x.
+ 4)(D2 + 4)
COS 2x
+ C2 sin 2x + C3Xcos 2x + qx
= D(D2
+ 4)2
= O.
Then
y = Cl
sin 2x
+ C5
Yc
and YP = Ax cos 2x
+ Bx
+ C.
sin 2x
4A sin 2x
+ 4B
cos 2x
Cl COS 2x
+ C2 sin 2x + "8x
+ "8'
sin 2x
+ 8D2)
= D5(D
+ 8)
= O.
Then
Yc
and Yp = Ax2
+ Bx3 + Cx4.
16A
=2
+ 9x
 6x2.
+ 6B
16A
=2
+ 24C =
48B
96C = 6.
Then A
11 ')
+ 1) to the
1)2(D + 1)(D3
Sx
Cl
1 4
 16x .
+D
 1) = D(D  1)3(D
+ 1)(D2 + 1)
Then
y=cle
+C2COsx+c3smx
+C4+cSe
Yc
142
x
+C6xe
2 x
+C7X e
= O.
Exercises 4.5
and
YP
Yp
4D = 1
2C
+ 4D = O
4B = 1
A=
Then A = 7, B
1/4, C
y=qe
= 1/2,
x
and D
= 1/4,
7.
.
+c2COsx+c3smx7+e
1 x
"2xe
+x
e.
Then
Yc
and
YP = A
+ Bx + Cx3ex.
Substituting
YP
A+3B=16
B=
1
6C = 1.
Then A = 13, B = 1, and
e = 1/6,
62. Writing (eX + ex)2 = 2 + e2x + e2x and applying D(D  2)(D
+ 2) to
obtain
Then
Yc
143
Exercises 4.5
and YP = A
+ Bxe2x + Ce2x.
Substituting.p,
2A
+ 9Be2x
= 2 + e2x + e2x.
 20Ce2x
Then
Yc
and YP = 'Ax2
+ Bx2ex.
= 1/2 and
= 1/2. The
+ 2Bex
= 1 + eX.
general solution is
Then
Yc
and YP = Ax2
+ 24C)
+ 16De2x
= 5x2  e2x.
+ 24C
=O
24B
=O
48C
=5
16D = 1.
Then A
y = q + C2X + C3e
2x
+ C4e
144
5
 16x
5
 48x
1 2x
 16xe
Exercises 4.5
+ C2e8x.
cle8x
A.
Substituting Yp into the differential equation we obtain 64A = 16. Thus A = 1/4 and
y = Cle
8x
+ C2e8x
1
 4
+ C2 = "4
ci
8Cl  8C2 = O.
Thus e = C2 = 5/8 and
5
y = _e8x
8
5
+ _e8x
8
+ C2ex.
1
4
+ 2B) + 2Bx
+ Bx2.
= x. Thus A = 1 and
B = 1/2, and
1
= Cl + C2ex  x + x
2
y' = C2ex  1 + x.
Y
= 1.
C2
Thus
ci
x
1 2
x + "2x
.
B= 1/10, and
Y
1
= 5c2e5x + 9  x.
25
25
+ C2 = O
41
5C2 = 125
145
10
x2
we find yp
= 2+x.
= Ax+Bx2.
Thus A = 9/25
Exercises 4.5
Thus CI = 41/125
41
5x
= 125 + 125e
9
1 2
+ 25x  10x .
Using D 
:2 to annihilate
lOe2x we find
and
+ C2= 4
6C2
"2.
5
+ _e6x
28
5
+ _e2x
4
 4sinx
we find Yp = Axcosx
+ Bxsinx
+ Ccos2x
+ Dsin2x.
= 2, B = O, C
= 8/3,
y
and D
= 8 cos 2x
 4 sin x. Thus
= O, and
= CIcos x + C2sin x +
y ,. =csmx+c2coSX+
2x cos x  ~ cos 2x
2 cosx
2xsmx+3sm
16 . 2x.
c2+=1
CI  7f = O.
Thus Cl =
+t:
70. The complementary function is Yc = Cl + C2ex + C3xex. Using D(D  1)2 to annihilate
xe" + 5 we find Yp = Ax + Bx2ex + Cx3ex.
146
Exercises 4.5
we obtain A + (2B + 6C)eX + 6Cxex = xeX + 5. Thus A = 5, B = 1/2,
y
= Cl +
1
1
c2ex + C3xex + 5x  _x2ex + _x3ex
C2+ C3+ 5 = 2
Thus q
= 8, C2=
71. The complementary function is Yc = e2x(cl cos 2x + C2sin 2x). Using D4 to annihilate x3 we
3
32x+
3
1
16x2+gX3
3
3
3
y' = e2x [cl(2cos2x  2sin2x) + c2(2cos2x + 2sin2x)] + 32 + gX + gX2.
The initial conditions imply
and
y
3
64
3
3
+ _x2
32
16
= e2X(2cos2x   sin2x) + x
1
+ _x3.
8
Using D2(D  1) to annihilate
x + eX we find yp = Ax3 + Bx4 + Cxe". Substituting yp into the differential equation we obtain
147
Exercses 4.5
(6A + 24B)  24Bx + CeX = x + eX. Thus A = 1/6, B = 1/24, and C = 1, and
y = Cl
I
x
12
13
y = C2 + 2C3X + C4e  2'x  '6x
12
y "2= c3 + C4ex  x  2'x
24.
+ xe x
+ ex + xe x
+ 2xe + xe x .
2
Thus
Cl
= 2, C2 = 1, C3 = O, and
C4
+ C4 = O.
2, and
x
y = 2 + x  2e 
'61x 3 
1 4
24 x
+ xe x .
73. To see in this case that the factors of L do not commute consider the operators
and (D
+ 4)(xD 
(xD  1)(D + 4)
+ 4xDx  Dx  4x
= xD2x
= x(O)
+ 4x(1) 
1
4x = 1
and
(D + 4)(x 1  x) = O.
Exercises 4.6
The particular solution, Yp = UlYl + U2Y2, in the following problems can take on a variety of forms,
especially where trigonometric functions are involved. The validity of a particular form can best be
checked by substituting it back into the differential equation.
148
Exercses 4.6
1. The auxiliary equation is m2
+ 1 = O,
so Yc
W=
Identifying
f (x)
= Cl COS X + C2 sin x
cosx
sinx
cosx
 sinx
and
I =1.
= sec x we obtain
I
Ul
=
U~
sin x sec x
=
cos x sec x
tanx
1.
Then
Ul
= ln I cos z],
U2
= x, and
y = Cl
COS X
+ 1 = O, so Yc = Cl COS X + C2 sin x
W=
Identifying
f (x)
xl
cos
I sinx
sin
cosx
and
l.
= tan x we obtain
l.
ul
I
U2
= smxtanx
cos2 X _. 1
= cosx  secx
cosx
= smx.
+ 1 = O, so Yc =
W=
Identifying
f (x)
= sin x
Cl COS X
+ e2
sin x and
cosx
sinx =1.
cos z
I sinx
we obtain
U~
I
U2
Then
=  sin2 x
.
= cos x sin x.
1
1
u 1 =  sin 2x   x
4
2
U2
1.
1
sm x cos x   x
2
2
=
2"1 cos2 x.
and
Y = el cos x
= el cos x
.
1.
2
+ C2 sm
x +  sm x cos x
222
+ e2 sin
1
x  2"xcos x.
149
1
12.
 x cos x   cos x sm x
4.6
Exercises
1 = O, so Yc
cosx
w=
Identifying
f (x)
= sec x tan
= Cl COSX + C2sin x
=  tan2
=  sinx(secxtanx)
U2 = cosx(secxtanx)
Ul
=x
 tanx,
U2
y = Cl COSX
+ C2sin x +
1 = O, so Yc
= tanx.
Cl COSX + C2sin x
cosx
w
sinx
cosx
I SInX
I =l.
=, 
SIn X COS x
U2 = cos3 x = cos x (1 Ul =
and
= cos2 x we obtain
Ul
Then
x = 1  sec2 x
and
= Cl cos X + C3sin x +
Identifying f(x)
I =l.
x we obtain
U~
Then
sinx
cosx
I sinx
and
sin 2 x) .
14
.2
(cos2 X
1.4
'3 SIn X
sin2 x) (cos2 X
1. 2
+ '3 SIn X.
ClCOSX+C2SInX+'3
1 = O, so Yc = Cl COSX
w=
Identifying f(x)
= sec2
cos x
+ C2 sin
sinx I =l.
cos x
I sinx
x we obtain
I
u 1
sinx
cos? x
U2 = sec x.
150
x and
sin2 x)
sin2 x
Exercises 4.6
Then
U =
1
cosx
= secx
U2 = In I sec x
+ tan
z]
and
y = c cos x
cos x
Identifying f(x)
+ C2 sin x
 cos x sec x
+ C2 sin x
 1 + sin x In I sec x
1 = 0, so Yc
= cex + c2ex
+ tan x l.
and
1 2x
1
+4
u = e
,
1 1 2x
u2 =  e
Then
U
1 2x
1
= e
+x
1 2x 1
U2 = e x
and
y = cie
= c3e
+ c2e x
1 x
 Se
1 (Xe
+ C4ex + x
1
+ xe
1 x
1 x
 Se  xe
 e X)
1 = 0, so Yc
ex
= 2.
ex
Identifying
f (x)
= sinh 2x we obtain
1
u~ = __e3x
1
+ _ex
,
1 x
1 3x
u2 = e
e
151
z]
Exercises 4.6
Then
Ul
1 3x
12
1 x
+e
= e
1 x
1 3x
e
e
U2 =
12
and
W =
Identifying f(x)
+ c2e2x
4 = O, so Yc = ce2x
I 2e2x
e2x
I = 4.
e2x
_2e2x
and
u2 = _x /4x.
Then
Ul
= lnlxl,
4
U2 = 
X dt
4 xo t
and
xQ
Identifying f(x)
+ c2e3x
9 = O, so Yc = cle3x
I ;::
Ul
=
I = 6.
_~~~3X
24 e
U2 = X
U2 =
1
 xe
and
152
and
~x.
6x
Then
> O.
Exercises
+ 3m + 2 =
W
Identifying f(x)
= 1/(1
+ eX)
(m
+ l)(m + 2) =
~:~x
_~~~2X
0, so Yc
= _e3x.
we obtain
eX
u,  __
1  1 + eX
e2x
Then
Ul
= ln(l + eX),
y
U2 = ln(l
eX
=  1 + eX = 1 + eX
u2
+ eX)
 e
 eX, and
+1=
w=
Identifying f(x)
= eX /
(1 + x2)
we obtain
,
u' 2 
Ul
= !In (1 + x2),
y
xexex
e2x (1+ x2)
+ 3m + 2 =
=
(m
z.
1_ex
ex
e2x
_2e2x
1
e3x
,
u2
Ul = 
eXeX
1
_e2x (1 + x2)  1 + x2'
vV
Then
+ x2
= 1
U2 = tan1 x, and
Identifying f(x)
I eX
ul = 
Then
(m  1)2
eX
= _e3x
+ c2e2x
153
+ ex
and
4.6
Exercises 4.6
+1=
(m  1)2
= jet
et
Identifying f(t)
= tanl t we obtain
I
=
Ul
tetettanlt
2t
e
Then
Ul
= ttan
1 + t2
= 2tan
1
t
t +"2
(1 + t2)].
t In
w=
Identifying f(t) = etInt
1) tanl
t
_et
t e t
tet
+ et
I e 2t

we obtain
Then
ul
= t1 2 In t + t1 2
2
U2 = z ln t  t
and
Y = qet
= Clet
1
1
2
+ c2tet  _t2et
In t + _t2et + t2et In t  t e:'
2
+ c2tet + ~t2et
In t  ~t2et
4'
+ 2m + 1 = O, so Yc = ex/2(
154
cl
cos x /2
Exercises 4.6
w=
ex/2 cos:!:
2
I _lex/2
222
Identifying f(x)
=
u2 = 
ex/2 sin(x/2)2.jX
/2
ex
= 4e
ex/2 cos(x/2)2JX
/2
ex
= 4e
Then
Ul
= 4
U2 = 4
y = ex / 2 ( Cl cos x
2
= jexsecx
+ 6 = O, so Yc = e" (Cl
6m
I e" cos x
+ C2 sin x)
e X smx
eX cos x
 e" sin x
= jx,
y = clex cosx
=
u2 =
and
2x
+ e" sin x = e
=  3 tan x
1
:3'
and
l'
+ 1 = (2m
w= tex/2~
cos X
+ 4ex/2
we obtain
I
U2
et/2Jt cos  dt
xo
2
eX cos x
ul
= j ln(cosx),
x/2 r:
x
v x cos .
2
w=
x/2 r:: . x
v x sm 2
+ C2 sin X)
Identifying f(x)
et/2Jt sin  dt
2
and
Ul
1 x
Ze .
Xo
Then
= 2.jX we obtain
u}
Identifying f(x)
ex/2 sin f2
I
cos :!: lex/2 sin:!:
lex/2
sin :!:
2
2
ex/2 .
I !ex/2
cosx
 1)2 = O, so Yc
xex/2
!xex/2 + ex/2
we obtain
155
+ :3xeX
sinx.
= Clex/2 + C2xex/2
I = e",
and
Exercises 4.6
Then
1 (
2)3/2
= 12 1  x
Ul
x r:; l.
1
U2 = V 1  x +  sin " x
and
I !eex/22 _!ee='? I = 1
2
.
x/
Identifying f(x)
U2 = xex/4+
x/
Then
Ul
= x2/8 and
eX/4. Thus
+ c2ex/2 + ~x2ex/2
y = clex/2
_ ~xex/2
+ ~ex/2
4
1
1
= C3ex/2 + C2ex/2 + _x2ex/2
__ xex/2
and
yf = ~c3eX/2 _ ~c2eX/2
+ ~x2ex/2
16
+ ~xex/2
8
_ ~ex/2.
4
1
1
 C2
224
C3
Thus
C3
= 3/4 and
C2
=1
C2
1

= O.
= 1/4, and
+m 
1 = (2m  l)(m
ex/2
W =
Identifying f(x)
I !ex/2
= (x + 1)/2 we obtain
156
+ 1) = O, so Yc = clex/2 + C2ex
and
Exercises 4.6
Then
Ul
_ex/2
U2
xe
1
3
Gx  2)
Thus
x 2
x/2 +C2e x
y=cle
and
,
x/2
y =Cle
x
c2e
.
C2 
2= 1
C2 
1
Cl
Thus
Cl
= 8/3 and
C2
= 1/3, and
y =
21. The auxiliary equation is m2
+ 2m 
8 x/2
1 x
3"e
+ 3"e
 x  2.
8 = (m X
2e2x  ex
+ 4) = O, so Yc = cle2x + C2e4x
2)(m
I=
e4x
 4 e 4x
w = 12e::
Identifying f(x)
O.
_6e2x.
we obtain
,
ul =
1 4x
1 3x
e
 e
1 3x
u2= e
6
1 2x
e
3
Then
Thus
y = cle2x
= cle2x
+ C2e4x
_ ~e2x
+ C2e4x
__ e2x
+ ~ex
12
and
y' = 2ee2x _ 4C2e4x
18
1
+ _ex
1
+ _e2x
2
157
__
1
ex.
9
+ ~ex
18
_ ~e2x
and
Exercises 4.6
The initial conditions imply
q
5
36 = 1
C2 
7
2q  4C2 + 18 = O.
Thus
q = 25/36
C2 = 4/9,
and
and
25
y = _e2x
36
Identifying f(x)
4
+ _e4x
9
__ 1e2x
4
+ 4 = (m  2)2
1
+ _ex
9
= 0, so Yc = qe2x
+ C2xe2x and
u = 6x2
 12x3
U2 = 12x2  6x.
Then
= 2x3
3x4
U2 = 4x3
3x2.
Ul
Thus
y
q e2x
+ C2xe2x + (2x3
3x4) e2x
+ (4x3
3x2) xe2x
3x2)
+ 2e2x
(x4 _ x3)
=1
ci
2q
Thus
q =
1 and
C2 =
+ C2 = O.
2, and
y = e2x _ 2xe2x
2x
1) .
= x1/2. From
Yl
+ .!:.y' +
x
(1 __4x21_) y = x1/2
= x1/2cosx
and Y2
158
= x1/2sinx
we compute
Exercises 4.6
Now
I
= COSX,
.
= smx
so
Ul
u; = cosx
so
U2 = sinx.
COSX
+ c2x1/2
sinx
+ x1/2
COS X
+ C2Xl/2
sin x
+ xl/2.
Ul
and
Thus
y = CIX1/2
=
ClX1/2
cos2 X
+ xl/2
sin2
y
and identify f(x)
= sec(lnx)jx2.
sec(lnx)
+ ~Y + x2Y
From
Yl
= x""'2'
= cos(lnx) and
sin(ln x)
cos(ln x)
w=
= sin(lnx) we compute
Y2
sin (In x)
cos(ln x)
1
x
Now
I
ul
=
tan(ln x)
x
so
Ul
= In I cos(lnx)'
and
I
u2
= 1 so
x
U2 = In x.
+m
= m(m2
1
w=
+ (lnx)
+ 1) = O, so Yc
cosx
sinx
 cosx
159
sin(lnx).
= Cl
+ C2 COSX + C3 sinx
sinx
cosx
sinx
=1.
and
Exercises
4.6
Identifying f(x)
= tanx we obt~in
cosx
sinx
 sinx
cosx
tan x
 cos x
 sin x
sinx
cosx
tanx
 sinx
=  sinx
cosx
u~ = W3 = O
O
sinx
O
tan x
cosx
= tanx
=  sin x tan x
cos2 X  1
= cos x  sec x.
cosx
Then
ln I cos z]
Ul
U2
= cosx
U3
and
y
+ sin 2 x
= C4
for
00
 sin x in I sec x
+ C2 COSX + C3 sinx
+ tan z]
ln I cos rl
>
sinx in I secx
+ tan
z]
cos 2x
= O 2sin2x
O 4 cos 2x
160
sin2x
2cos2x
4 sin 2x
= 8.
and
Exercses
Identifying
f (x) = sec 2x
we obtain
,1
u1
4.6
= W
1 =
8
1
8
cos 2x
sin 2x
2 sin 2x
2 cos 2x
sec 2x
4 cos 2x
4 sin 2x
sin2x
2 cos 2x
O sec 2x
,1
u3 = 8"W3
1
4
4 sin 2x
1
cos 2x
1
8" O 2sin2x
O 4 cos 2x
= 4 sec 2x
=  tan 2x.
sec 2x
Then
U1
1
= 8"In I sec 2x
U2
U3
1
= 8"lnlcos2xl
+ tan
2xl
'4x
and
Y = CI + C2 cos 2x
for
+t:
/4
<
2x I 
<.7f /4.
YPl
= A cos x
C2
sin 3x).
equation we get
(27A  6B)cosx
+ (6a+
Then
27A  6B = O and
6a
+ 27b = 15,
so A = ~ and B = 197 . Thus, YPl = (7 cos X + f.7 sin x. Next, we consider the differential equation
3y"  6y' + 30y, for which a particular solution YP2 can be found using variation of parameters. The
Wronskian is
W=
Identifying f(x)
eX sin 3x
I = 3e2x.
3ex cos 3x + eX sin 3x
eX cos 3x
, = '9sm
l.
3xtan 3x
u1
and
161
u2
l.
Exercses
4.6
Then
Ul
27
1
u') =  cos3x.
27
Thus
YP2
We consider first the differential equation y"  2y' + y = 4x2  3, which can be solved using
undetermined coefficients. Letting
= Ax2 + Bx +
YPl
equation we get
= 4x2  3.
4A + B = 0,
and
2A  2B +
e = 3,
so A = 4, B = 16, and e = 21. Thus, YPl = 4x2 + 16x + 21. Next we consider the differential
equation y"  2y' + y = x1ex, for which a particular solution YP2 can be found using variation of
parameters. The Wronskian is
w=
Identifying f(x)
= eX/x
we obtain u~
eX
xe"
eX xe" + eX
I = e2x
1 and u2 = l/x.
Then
Ul
= Yc +
YPl
YP2
29. The interval of definition for Problem 1 is (7r /2, 7r/2), for Problem 7 is (00,00),
for Problem 9
< lnx < 7r/2 or e1r/2 < x < e1r/2. The bounds on lnx are due to the presence of sec(lnx)
and
y"
162
Exercises 4.7
so
X4y"
Letting w
= u',
Thus, w = x5
=  dx
In Iwl
and
+ c.
= 5lnx
and u = tx4.
= O.
= x2x4
I x2 1/x2 I = _. .
2x
Identifying f(x)
U2 =
t lnx,
tam u =;jX
1/ X4 we obtai
_2/x3
5
an d' u2 = ;jX
1
Then u
i 4
I6x
an d
so
Y
1
=  x16
x 2  1 (ln x )x 2
4
4
1
=  x16
2  1_2
x In x.
4
C2
1
1
+ x2
  4x2
16x2
In x.
Exercises 4.7
+1=
+ l)(m
 2) = Oso that y
= CIXl + C2x2.
+ C2xl/2Inx.
+ C2x2+v'6.
is m2 + 2m  4 = O so that y = CjXI+v's + C2XIv's.
is 25m2 + 1 = Oso that y = Ccos (i In x) + C2(! In x)
1 = Oso that y
163
= xI/2
[CIcos
(V;
In x)
+ C2sin
(V;
lnx ) ].
Exercses 4.7
14. The auxiliary equation is m2  8m+ 41 = O so that y
= x4
[C1cos(51nx) + C2sin(51nx)]:
15. Assuming that y = xm and substituting into the differential equation we obtain
m(m  l)(m  2)  6 = m3
3m2 + 2m  6 = (m  3)(m2 + 2)
= O.
Thus
y
16. Assuming that y
m(m l)(m
 2) + m  1 = m3
3m2 + 3m  1 = (m  1)3 = O.
Thus
m(m1)(m2)(m
Xffi
3) +6m(m1)(m2)
+9m(m1)
Thus
y
= m(m
 5)
= O so that
Yc
= C1+
C2x5 and
x5 = 5x 4 .
W(l, x 5 ) = O1 5x4
Identifying f(x) = x3 we obtain
51515
y=C1+C2X
25x
+Sx
515
lnx=c1+c3X
+Sx
U2 = ilnx,
and
lnx.
20. The auxiliary equation is 2m2 + 3m + 1 = (2m + l)(m + 1) = O so that Yc = C1X1 + C2X1/2 and
W(x1
Identifying f(x) = ~
x1/2) =
1
X
_x2
we obtain u = x 
1/2
X
_~x3/2
I = _x5/2.
1
2
1
1 2 2 2 2
= C1x1 + C2x1/2 + 2x
 3x + 5x  x
3
164
1
= C1x
1/2
1
1 2
+ C2x
 x + x .
6
15
4.7
Exercises
21. The auxiliary equation is m2  2m
+1=
(m  1)2
w (x, x ln x)
Identifying f(x)
and
= 2/x
we obtain
ul
= 21nx/x
y = C1X
Identifying f(x)
= x2ex we obtain
I1
= x
ul
x ln x = x.
1 + ln x
and U2
+ C2X ln x
+ 2 = (m
 x(ln x)2
 l)(m  2)
= x2ex
= z. Then
+ 2x(ln
Ul
U2
= 21nx,
x)2
= Oso that Yc
and U2 = xe".
= (lnx?,
Then
= ClX
Ul
+ C2x2
and
= _x2ex + 2xex
2ex,
+ 2m
y = Cl + C2X2
= m(m
and
+ 2)
= O, so that
y' = 2C2X3.
2C2 = 4.
Thus, Cl = 2, C2 = 2, and y
= 2  2x2.
20
y
= ClX2 + C2x4
+ 8 = (m
and
y'
 2)(m  4) = O, so that
2ClX
+ 4C2X3.
4
4Cl + 32c2 = O.
Thus, Cl = 16, C2 = 2, and y
= 16x2
20
2x4.
30
165
Exercses
4.7
equation
+ 1 = 0,
is m2
so that
y
3
+ C2sin(lnx)
y = C cos(lnx)
and
. (ln z )
y , = CI ~1 sm
x
The initial conditions
+ C21 cos(in
x
lOO"
50
x ).
3
+ 2sin(lnx).
y = cos(inx)
+ 4 = (m
is m2  4m
Y=Clx2+c2x2inx
and
 2)2 = 0, so that
y'=2CIX+C2(x+2xinx).
imply CI = 5 and C2 + 10
= 3.
20
30
= O so that
Yc = CI + c2in x and
15
W(l, ln z) = 11 in/x I = ~.
O 1 x
x
Identifying
f(x)
UI = tx2  !x2inx,
equation
so that Yc = CIX2
UI
f(x)
u2 = x.
10
Then
+8
in x = c
C2+ ~ =
12
+ C2in x +  x
4
!. Thus,
= (m  2)(m  4)
CI =
~, C2 = 1,
0,
y
0.05
and
= I x2
2x
X43]
4x
= 8x4 we obtain
= x4, U2 = 2x2,
12
+ x
t = 1 and
is m2  6m
+ C2x4
W
Identifying
  x in x
+ t x2.
in x
1212
+ x
imply CI +
= ~
ln z and
U2 = ~x2, and
y = CI + C2in x
and y
u~ = +z
1 we obtain
= 2x5.
u~ = 4x3
and y = CX2
and
+ C2X4 + x6.
u2 = 4x.
Then
166
1 x
1
imply
Exercises 4.7
1
Cl
4
1
1
=16
64
1
3
Cl + 2"C2 = 16 .
1
1
d
Th us Cl = 16'
C2= "2' an y
+ C2
1 2 1 4
= l6X
"2X
+ X 6.
d2y
dt2
+8
dy
dt  20y = O.
+ 10)(m
 2) = O so that
d2y
dy
dt2  10 dt
+ 25y
= O.
d2y
dy
dt2 + 9+ 8y
dt
The auxiliary equation is m2
= e2t
+ 9 + 8 = (m + l)(m + 8)
= O so that
+C2et+30e
1 2t
=C1x
1
+C2X
Yc = clet
= e2t, so that A
8
1
+30x.
+ C2e8t. Using
= 1/30 and
+6
+ 6y = 2t.
+ C2e3t. Using
= 2t, so that
At
d2y
dy
dt2  4 dt
+ 13y
167
=4
+ 3e .
Exercses
4. 7
The auxiliary equation is m 2 4m + 13 = Oso that Yc = e2t (ci cos 3t + C2sin 3t). Using undetermined
coefficients we try Yp = A+Bet.
and
.
Y=e2t(clcos3t+c2sm3t)+
=X2[Clcos(3lnx)
4
3
13 + 10et
+c2sin(3lnx)1
4
3
+ 13 + 10x.
34. From
it follows that
dy
+ 11 dt
 6y
= 3 + 3t.
= (m1)(m2)(m3)
C3e3t. Using undetermined coefficients we try YP = A + Bt. This leads to (11B  6A)  6Bt = 3 + 3t,
so that A
168
Exercises 4.7
In the next two problems we use the substitution t = x since the initial conditions are on the interval
(00, O). In this case
dy
dy dx
dt = dx dt =
dy
dx
and
+ c2t1/21n
y(t)
'
+ 1 = (2m
I
t=l
=2,
 1)2 = O, so that
+ C2
y' = ~CICl/2
and
It=l = 4.
y'(t)
(Cl/2 + ~cl/2In t) .
= 2tl/2  5tl/2ln t
< O.
y(t)
It=2 =
y'(t)
8,
t=2
= O.
+ 6 = (m  2)(m  3) = O, so that
= clt2 + C2t3 and y' = 2clt + 3c2t2.
4Cl + 12c2 = O
from which we find c
= 6 and
37. Letting u = x
+ 2 we obtain
ddy
x
d2y
dx2
= 6t2 
2t3
= 6x2 + 2x3,
< O.
d2y du
= du2 dx
d2y
du2(1)
d2y
du2
= c cos[ ln(x
169
Exercises
4.7
38. If 1  i is a root of the auxiliary equation then so is 1 + i, and the auxiliary equation is
= m3 
1, c
4m2 + 6m  4
= O.
+bm(m1)
3, and d
equation is
+ 3xy'
X3y'"  x2y"
39. For x2y"
 4y = O.
= O the auxiliary equation is m(m  1) = O and the general solution is y = Cl + C2X. The
y'(O) = Yl implies Cl = Yl. Thus, the initial conditions are satisfied for YO = O and for all real values
of Yl.
For x2y"  4xy' + 6y = O the auxiliary equation is m2  5m + 6
v'xcos(lnx)
for all negative integers and the graph has infinitely many xintercepts in (0,0.5).
41. The auxiliary equation is 2m(m  l)(m  2)  10.98m(m  1) + 8.5m + 1.3
= O, so that ml
= C1XO.053299
+ C2X1.81164 + C3x6.73l66.
2.22291i.
1.40819 and
43. The auxiliary equation is m(m  l)(m  2)(m  3) + 6m(m  l)(m  2) + 3m(m 1)  3m + 4
so that ml
= O,
+ C3Xv'2+ c4xv'2ln x.
so that ml = m2
+33m(m1)
105m+ 169 = O,
is
Y
Exercises 4.8
Exercises 4.8
1. From
Dx = 2x  y and Dy = x we obtain y
= 2x 
Then
X
= Clet +
)t + C2tet .
C2tet an d Y = (Cl  C2e
5t
Y = "7cle
3t
 C2e
x = Cl COS t + C2sin t + 1 + t
and
y = Clsin t  C2cos t + t  1.
4. Frorn Dx  4y
= 1 and
x + Dy
= 2 we obtain y = iDx
..t,
Dy
= iD2x,
= 2.
Then
x = Cl COS t + C2sin t + 2
and
1
1.
1.
1
y = C2cost  Cl smt  Cl smt  .
i  i(D
Dx = i(D2
and
= Cl COS /5 t + C2sin /5 t  ~
( 32
/5)
x = Cl  C2
3
cosv'5t+
171
(/5
Cl
3
2).
 C2 smv'5t+
3
.
5
+ 2)y,
Exercises 4.8
(D2
+ 4)(D
+ 2)x
 2)(D
= 3et.
Then
and
.
(D  5)(D2
+ 5)x + Dy = O and
+ 4)y = O. Then
2t
C2 SIn
+ l)x + (D
(D
 4)y
1 t
Se
.
x = cle5t
+ C2 cos 2t + C3 sin 2t
y = c4e5t
and
Substituting into (D
(6e
so that
C4
+ l)x + (D 
4)y = Ogives
6Cl,
= ~C3,
C5
C6
+ D2y =
= ~C2, and
t
y = 6cle"
9. Frorn Dx
+ (2C2 + C3 
+ 2'C3 cos 2t
+ l)x + (D
e3t and (D
 l)y
+ C2 SIn. t + C3 cos t 
and
x
Substituting into (D
+ l)x + (D
(C4 so that
C4
ci)
4 3t
15e
17
C4
+ (C5 
C6 
+ (C6 + C5 + C2 
C3 
C2) sin t
.
cos t + C3 SIn t
C3) cos t
=O
C2
+ (D + 3)y =
and
172
17 3t
15e
2 we obtain D(D
+ l)(D + 3)x = 1 + 3t
and
Exercises
Substituting into (D
4.8
gives
2
and
Y = q
t
C2e  3c3e
3t
+ t  "21t 2 .
(D2  l)x  y = O and (D  l)x + Dy = Owe obtain y = (D2  l)x, Dy = (D3  D)x, and
(D  1)(D2 + D + l)x = O. Then
11. Frorn
J3 + C3sin
. =:'
J3]
= cie t + et/2 [C2cos 2t
and
y =
(~C2
2
+ l)(D + l)y
= 1 and
(D1)x+Dy
= 1 we obtain
(2D+1)(D1)(D+1)x
= 1
= 2. Then
y = C4et/2 + C5et  2 .
Substituting into (D  l)x
+ Dy
= 1 gives
+ (2C2
 c5)et
=O
C5 = 2C2, and
y =  3Cle t/2  2 C2et  2 .
= et and
(D1)x+Dy
= 5et we obtain
Dy = (52D)x+
and (4D)x
4et.
Then
and Dy
= et and
4t +C2+5e.
= et. Then
and
173
Exercises
4.8
y = Cl  C2 C2t C3et  tet + et.
15. Multiplying the first equation by D + 1 and the second equation by D2 + 1 and subtracting we
obtain (D4  D2)x
1. Then
X
Multiplying the first equation by D + 1 and subtracting we obtain D2(D + 1)y = 1. Then
12
so that C5= Cl  C2+ 2, C6= C2+ 1, and C7= C4. The solution of the system is
.X
= Cl + C2t +
12
t
+ Dy
we obtain x
= Dy,
D2x
17. From Dx
1
2
x = (C2+ C3)e  t sin t + (C2 C3)e  t cos t + 5" sin t  5" cos t.
= y, Dy
= z. and
. Tt
J3 + C3cos TtJ3 ] ,
x = cj et + e t/2 [C2sin
y = Clet +
and
z=qe
t+
(1 J3)
(1"2c2+Tc3J3)
C2  C3
22
. t
J3 +
e t/2 sm
2
. J3
e t/2 smTt+
1)
(J3
TC2"2C3 1)
(J3
C2
22
t/2
 C3 ecost
J3
2'
J3
e t/2 cosTt.
+ C2COS t + C3sin t +
3 2t
.
"2Cle  C2cost  C3smt,
and
174
et,
Exercises
=
4.8
+ C2 sin t.
Dx  6y = O
x  Dy + z = O
x
+y
 Dz = O.
Multiplying the second equation by D and adding to the third equation we obtain
(D + l)x  (D2  l)y = O. Eliminating y between this equation and Dx  6y = Owe find
(D3
Thus
x = clet
+ C2e2t + C3e3t,
and, successively substituting into the first and second equations, we get
1 t
1 2t
Y = "6cle  3"c2e
5
6
t
Z = Cle
1
3
C2e
2t
3t
+ "2c3e
1 3t
+ C3e
2
(D+l)xz=O
(D
+ l)y
(D2+Dl)z
= O. Eliminating
+ 1 and
z =O
+ Dz
x  y
= O.
+ l)x +
= Owe find
D(D+l)2x
so that Dy
= _(D2 + 5D)x.
= O.
Thus
x
Cl
+ C2et + c3tet,
and, successively substituting into the first and third equations, we get
y = Cl
+ (C2 
+ c3tet
c3)et
= Oand 4x
and
Y
=
(2 Cl
+ C2) e 3t
175
2 C2t e 3t
Exercises
Using x(1)
4.8
O and y(l)
1 we obtain
or
Cl
2Cl
Thus
Cl
= e3 and
22. From Dx  y
Then _!(D2
C2
+ C2 = O
+ 3C2 =
_e3.
= 1 and
 2D)y
3x
+ (D
 2)y
= e3t+3
_ te3t+3
_e3t+3
= Owe obtain
y = et
( Cl COS
+ 2te3t+3.
x = !(D
 2D)y.
= 3. Thus
v2 t + C2 sin v2 t) + 1
and
x = ~et
[( ci 
+1
=O
~ (c  v2 C2) + ~ = O.
Thus
Cl
= 1 and
C2
V2/2. The
='
= et
(
cos v2 t
V2.)sin v2 t + 1.
+2
23. Equating Newton's law with the net forces in the x and ydirections gives m d2X2 = Oand m d2Y2=
dt
dt
mg, respectively. From mD2x = Owe obtain x(t) = clt+C2, and frorn mD2y = mg or D2y = g
we obtain y(t) = _~gt2
+ C3t + ca.
+:
dt2
.
1~
mg  ksme = mg  k = mg
v dt
176
Icl.~
dt
Exercises 4.9
From mD2x + IclDx = Owe have D(mD + Icl)x = Oso that (mD + Icl)x = Cl. This is a firstorder
linear equation. An integrating factor is ellcldt/melclt/m so that
!!_ [eic1t/mx]= c
dt
elclt/m
1
= c3+c2elclt/m. From
(mD2 + IclD)y = mg we have D(mD + lel)y = mg so that (mD + Icl)y = mgt
firstorder linear equation with integrating factor elclt/m. Thus
d
 [elclt/my]= (mgt
dt
and
+ Cl This is a
+ cl)elclt/m
Exercises 4.9
= eX, so
1. We have y~ = yI
Y2 = 
sin x and y~ =
cosx, so
(y~) 2 =
However, if y
(y")2
=1=
f = cos2x = y~.
qYl + C2Y2, we have (y")2 = (cleX  C2cos x)2 and y2 = (ci eX + C2cos xf
y~
= yI = O, so
YlYl"
Y2 = 2x
1 . O = O = 1()2
O
2
However, if y
= 21(Yl')2 .
and y~ = 2, so
Y2Y2" = x 2(2 )
~(y/)2
cos x
y2.
2. We have
Also,
(
= ~[Cl . O+
= 2x 2 = '21(2x )2 = '21(Y2')2
c2(2x)]2
= 2c~x2.
Thus yy"
=1=
~(y/)2.
177
Thus
Exercses 4.9
3. Let u :::;:
y' so that u' = y", The equation becomes u' = u  1 which is separable. Thus
du
2
+1
du
l+u
2
= dx ==> tan
u = x + Cl ==>
= tan(x
= In 1COS(Cl x)1 + c2
cj )
= y' so that u' = y", The equation becomes x2u' + u2 = O. Separating variables we obtain
5. Let u
du
u2
dx
=  x2 ==>
;
= ;; + Cl =
ClX+ 1
X
1
1
Y = 2" In IClx + 11 x
cl
Cl
1 (
==>
=  Cl
+ l/Cl
1 (1
= Cl qx + 1  1
+ C2
,
"du
.
6. Let u = y so that y = u dy' The equation becomes (y
+ 1)u du
dy =
u . Separating variables we
obtain
du
dy
y+1
+ 1)
dy
==> = Cldx
y +1
C3eCx.
3
= y , so that y "du= u dy' The equation becomes u du
dy + 2yu = O. Separating variables we
7. Let u
obtain
du
2"
u
+ 2y dy = O==>
+ y2 = C
==> u=
==>
8. Let u
= y , so that
"du
= u dy'
dy
du =  ==> u
y2
q
If Cl = O, then y dy
= dx
y2
1
Y
) dy
qy  1
,1
==>y=
+ ci
y2
+ ci
"3 y + qy = x + C2
.
2 du
The equaton becomes y u dy
=  + Cl ==> Y
==> _!:_ (1 +
= u.
qy  1
Y
==>
dy
Y
ClY 1
= dx
h2 = x
178
+ c2.
cl
Exercises 4.9
9. (a)
Iy
l.
10
(b ) Let u
du
= y ' so that y 11 = u du
dy . The equation becomes u dy + yu = O. Separating variables we
obtain
When x
= 0, y =
1 and y'
dy
1
=y
dx
When x
= 0, y
=
1
1 so 1
= ~ + Cl and
dy
1
=  dx
===?

===?
Y = tan ( ~x
y2
+1
= 1 so 1 = tan e, and C2
Cl
tan
===?
+ C2)
= ~. Then
1
y = 2x
+ C2
179
Exercises 4.9
10. Let u = y' so that u' = y". The equation becomes (u')2
which results in u' =
~.
+ u2 =
To solve u' = ~
we
separate variables:
du
Jl="U2
= dx
27t
27t
7r ,
j3
(7r)
j3
"67r .
When x =
7r
"2'
(x ~)
(7r
7r)
(x  ~) + e2.
==> y =  cos
= ~
(x  ~).
we separate variables:
du
v'l=U2 = dx
==> cosl u =
+ el
7r ,
j3
(7r)
j3
= cos (x
 i)
==> y
el
7r
=  3". Thus
= sin (x
. (7r
= 7r
"2 ' y = 2"1 ' so 2"1 = sm
"2  3"7r) +e2 = 2"1+cz and
When x
problem is y = sin
 i) + e2
e2
..
..
(x  i)1
11. Let u = y' so that u' = y". The equation becomes u'  u = _u3, which is Bernoulli. Using
W
=U
.,

2
2
.
.
. 2
we obtam  + w =   _An mtegratmg factor lS x , so
dx
x
x
d 2 1
2
2
el
dx[x w = 2x ==> x w = x + c ==> w = 1 + x2
.
dw
2
==>
==>
dy
dx
ci
= 1 + 2"
x
Vel  x2
==> u=
Vel  x2
==> y = 
J el 
x2
+ e2
180
Exercses
12. Let u
dw
su bstituti
itution w = u 1 we obtai
tain d
d
[xw]
dx
= x
==> w
1
= X+C
2
. factor IS
. x, so
= .1 An mtegratmg
1
+ w
1
1
==> x
u
4.9
Cl 
x2
2x
==> u =
2x
Cl 
x2
==> y
= In ICl
x2 +C2.
In Problems 1316 the thinner curve is obtained using a numerical solver, while the thicker curve is the
graph of the Taylor polynomiaJ.
13. We look for a solution of the form
1
1
y(x) = y(O) + y'(O) + 2Y"(0) + 3TY"'(O)
From y"(x)
+ 4Ty(4)
(x)
+ 5iy(5)
(x).
40
= x + y2 we compute
y'''(x)
y(4)(x)
= 1 + 2yy'
= 2yy" + 2(y')2
y(5)(x) = 2yy'"
Using y(O)
+ 6y'y".
y"'(O) = 3,
y(4)(0) = 4,
y(5)(0) = 12.
An approximate solution is
y(x)
Frorn y"(x)
1
10
1
+ y(4)
4.
(x)
1 (
+ 51Y
5)(x).
= 1  y2 we compute
y"'(x) = 2yy'
Using y(O)
y(4)(x) = 2yy"
_ 2(y')2
y(5)(x) = 2yy'"
 6y'y".
= 2 and y'(O)
y"(O) = 3,
= 3 we find
y"'(O) = 12,
5
y(4)(0)
= 6,
y(5)(0)
102.
10
181
Exercises 4.9
An approximate solution is
3
y(X) = 2 + 3x  x
2
15. We look for a solution of the form
1
1
y(x) = y(O) + y'(O) + "2Y"(O) + 3iylll(O)
3
1
 2x  x
4
+ x17
20
.
y
4T1y(4) (x)
5T1y(5)(x).
= x2 + y2  2y' we compute
From y"(X)
+ 2yy'
ylll(X) = 2x
= 1 and y'(O)
y"(O)
= 1,
+ 2yy"
 2ylll
20
 2y(4l.
= 1 we find
y'"(O) = 4,
30
+ 2yy'"
40
 2y"
y(4)(x) = 2 + 2(y')2
Using y(O)
10
= 6,
y(4)(0)
y(5)(0)
= 14.
An approximate solution is
12231475
y(x) = 1 + x  x + x
+ x
 x
60
= y(O) +y' (O) + ~y" (O) + ~ylll (O) + ~y(4) (x) + ~y(5) (x) + ~y(6) (x).
From y"(X)
ylll(X)
y(4)(x)
2
3!
eY we compute
4!
5!
61
= eYy'
= eY(y')2 + eYy"
y(5)(x) = eY(y')3
+ 3eYy'y" + eYylll
y(6)(x) = eY(y')4
(O) = 1,
10
2
4
y 111(0) =,1
An approximate solution is
y(x)
(1 + u2) 3/2
= u' or
x
+ x121314x1516+ x
2
[1 + (y')2f/2
(1 + u2) 3/2
= y".
12
+ x
24
Let u
+ x
45
182
tan
Exercises 4.9
we have
(
du
1 + u2)3/2 = dx
==:;.
==:;.
18. Let u =
dx
di
==:;.
==:;.
y'
sec2 e
(
)3/'> de = x
e 
1 + tan2
cos e de = x
y'
1+
sin e = x
==:;.
==:;.
sec2 e
sec3 e de = x
==:;.
~
[]
(y')2
=x
x2
 x
vIl=X2
(for x
> O)
==:;.
=~
du
_k2
The equation becomes u dx = ~.
d2x
du
so that dt2 = u dx
Separating variables we
obtain
k2
udu=2dx
x
When t
==:;.
u
2
k2
x
=+c
k2
k2
Xo
Xo
= k2
(~
2.)
Xo
Then
dx
~/xoo
 x
=kv2
dt
xXo
and
 ~
V~
==:;.
Po
kV"2
t=~
J/
x
xox
t =  1 ro [x
k
[V
=1 ro
x(xox)+tan
Xo
2
Xo  2x
2Jx(xo  x)
dx.
{x!;]
Xo  x
1.
19.
2
zp'
20 '
2
d2x
For dt2
xl = o
+ sin x =
2
xl
183
Xl.
Xl
O. The
Exercises 4.9
lO'
1
d2x
For dt2
dx
dt
+ sin x = Othe
lO'
10 '
1
xl = O
1
xl= I
x l = 2.5
= cos x,
From (y")2  y2 = O we have y" = y, which can be treated as two linear equations. Since linear
combinations of solutions of linear homogeneous differential equations are also solutions, we see that
y
clex
+ C2ex and
C3 cos x
combinations that involve both exponential and trigonometric functions will not be solutions since
the differential equation is not linear and each type of function satisfies a different linear differential
equation that is part of the original dfferential equation.
21. Letting u
f1+2
1i u~,
VI du+ u2
du _
dx  V
and
 d
 x,
sinhl u = x
+ CI.
Then
u = y"
= sinh(x + CI),
y' = cosh(x
+ CI) + C2,
and
=O
2. Since Yc
 Y = 1 + e" is yp
= A + Bxe".
3. True
4. True
5. They are linearly independent over (00,00)
6. (a) Since h(x)
= 21nx
+ 1 is not
= fz(x),
dependent.
(e) Since h(x)
(f) Since 2x is not a constant multiple of 2, the functions are linearly independent.
(g) Since 3(x2)
(h) Since xex+l
+ 2(1 + 0(4x
+ y"
 17y'
+ 5)(m
 1) = m3
+ m2 
17m
+ 15 = O,so the
differential
+ 15y = O.
+ 5)(m
 1) = m3
+ d = m3 + (b 
+ m2 
17m
+ 15 =
3)m2
+ (c  b + 2)m + d = O.
O, we have b  3 = 1, c  b + 2 = 17,
and d = 15. Thus, b = 4 and c = 15, so the differential equation is ylll + 4y" 15y'
+ 15y
= O.
8. Variation of pararneters will work for all choices of g(x), although the integral involved may not
always be able to be expressed in terms of elementary functions.
coefficients will work for the functions in (b), (c), and (e).
9. From m2  2m  2 = O we obtain m = 1 y'3 so that
y
10. From 2m2
11. From m3
+ 2m + 3 = Owe obtain
+ 10m2 + 25m
Cl e(l+V3)x
+ c2e(lv'3)x.
m = 1, m
+ e3x/2
185
V;
+ C3 sin
V;
x) .
so that
+ 3m3 + 2m2 + 6m 
4 = O we obtain m
y = crex/2
VTI
2
. VTI)
2
+ C2sin x
+ 5) = O. Then
Yc
and Yp
= A + Bx + Cx2 + Dx3.
(5A  3B
+ 2C) + (5B
+ 6D)x + (5C
 6C
 9D)x2
(VTI
VTI)
+ 5Dx3 =
36/25, and D
36
+ x46 + x
125
+ 4x3.
2x
25
4
+ x
5
+ 1) = (D
Then
Yc
and YP = Ax2ex
+ Bx3ex + Cx4ex.
12Cx2ex
Equating coefficients gives A = O, B
= O,
Y = C 1eX + C2xe"
+ 1) to
+ 1)(D3
D(D2
 5D2
 2)(D  3)
= O.
Then
Yc
and YP = Ax
+B
cos x
+ C sin x.
6A
+ (5B + 5C)
cosx
+ (5B
= 1/5, and C
2x
3x
+ C3e +"3x
186
1 .
 1)5 = O.
Yc
y = Cl
19. The auxiliary equation is m2
+ C2X + C3ex
 3x2.
and
w=
Identifying
eX sinx
I e" cos x
+ eX sin x
=
,
(eX sinx)(eX tan x)
u2 =
2
e X
Ul
eX cosx
exsinx+excosx
2x
= e
=
cosx,
U2
y = Cl eX sin x
=
Cl eX
sin x
= sinx
In
= Slnx
sin2x
= cosx = cosx
and
+ C2ex cos x
 eX cos x In I sec x
1 = 0, so Yc = qeX
+ tan x l.
+ C2ex
e X
ex
Identifying f(x)
 secx.
+ C2ex cos x  e" sin x cos x + eX sin x cos x  eX cos x In I sec x + tan z]
= 2ex /(eX
+ eX)
eX
1
and
I = 2.
we obtain
1
u' 
+ ex
eX
 0' 1 + e2x
,
e2x
e3x
X
eX
u2 =  eX + eX =  1 + e2x = e + 1 + e2x'
Then
Ul
Then
= 6. Equating coefficients
m  1=
y
so that
= qXl/2 + C2Xl/3.
187
3
+C2X
3
Inx+4:x
(m  2)(m  3)
= Oand
= Clx+c2xlnx+
1
4:x3.
When w
i
a, Yp =
When w
i a, YP = Al<t and
+ c2ewt + Aeat.
26. (a) If y = sinx is a solution then so is y = cosx and m2 + 1 is a factor of the auxiliary equation
m4 + 2m3 + 11m2 + 2m + 10 = o. Dividing by m2 + 1 we get m2 + 2m + lO, which has roots
1
differential equation is
y
= Clsinhx
+ C2coshx
+ C3Xsinhx + C4Xcosh x.
(b ) Since both sinh x and x sinh x are solutions of the associated homogeneous differential equation,
a particular solution of y(4)  2y" + y = sinhx has the form YP = Ax2 sinhx + Bx2 coshx.
188
yi
= _x2  2x
+ x2 + 2x = O, and
Yl
= x is
a solution of the associated homogeneous equation. Using the method of reduction of order, we let
y
CX
+ C2xex.
= eX is a solution of
that u"  u' = 1. This differential equation has a particular solution of the form Ax. Substituting,
we find A = 1, so a particular solution of the original differential equation is YP = _x2 and the
general solution is y
= cx
+ C2xex
 x2.
= e7r and
Cl
+ 2m + 1 =
C2
= O. Thus, y = eX7r
+ C2 sin x).
Setting
cosx.
(m
+ 1) =
O so that m
1 and y
clex
+ c2ex.
= O, and e = !.
Cl
+ C2ex 
.
"21 sin
x.
x 
Cl 
Cl
and
li
C2 
i.
3
= 3.
2
The solution of the initialvalue problem is

13 x 5 x
1 .
= e
 e  x   smx.
C2
+ C2 = 2
442
cosx
sin x =1.
cosx
sinx
ul
= smxsec
x =
sin x
3COS X
189
1
1
= sec
2 cos' X
2
= 
u2 = tanx.
Thus
Y=
cr cos x + C2 sin
X
= Cl
=
and
2' sec x +
COS X
C3 COS X
,.
Yp =
C3
+ C2 sin X
1
COS2 X
cos x
sin x
x.
C3
+ 2'
=1
C2 =.
Thus
C3
= C2 = 1/2 and
Y =
33. Let u
1.
= y' so that u' = y". The equation becomes u dx = 4x. Separating variables we obtain
u du
When x
= 4x dx
===:;,.
1
_u2 = 2x2 +
2
= 1, y' = u = 2, so 4 = 4 + C2 and
C2
C1 ===:;,.
u2 = 4x2
+ C2.
= O. Then
or
dy = 2x
dx
=y
so that y
"du
= u .dy
= x2 + 4.
.
du
The equation becomes 2u dy
2u dy = 3y2 dy
===:;,.
190
= 3y 2 .
u2 = y3 + Cl.
o, y
= 1 so
= y3
==}
(~~)
==}
_2yl/2
u2
When x
1 = 1 + Cl and Cl
= 1 and y' = u
= O, y = 1, so 1
y3
==}
=X
+ C2
2 =
~~
= O. Then
= y3/2
==}
==}
Y = (x
y3/2 dy
= dx
+ C2)2 .
4
4
4
Note
= 2" ==} C2 = 2. Thus, y = (x + 2)2 and y =
(x
2)2 .
c2
4
,
8
+ 2)2 , y =  (x + 2)3 and y'(O) = 1 =11. Thus, the solution of the
+ 64m3 + 59m2
 23m 12
1/3,
3
1
4e  C2  C3
+ C4
232
16c
64cl
=2
+ C2 + 9C3 + 4C4 =
27
8
 C2'
1
27
C3
+ C4
= O.
8
191
36.
Consider xy"
+ y' =
'f
+ y'
= m(m  l)xml
1
+ mxml
= m2x.
2
+ y'
+ y'
= O is Yc = Cl + C2lnx. To find a
IX
The Wronskian is
xl/2Inx
= y'Xlnx
l/x
so that
ul
= x3/2(
5
we obtain
ul =
Then yp
q
~~:I=~
w=l~
3
= x 3/2(2Inx
393
~ In x  ~)  ~x3/2In x
is y = Cl + c2Inx  ~x3/2
u; =
and
4)
 
=  ~x3/2
and
u2
_xl/2
= VI,
s.]
= x2
3/2 .
The initial conditions are y(l) = O and y'(l) = O. These imply that
+ (D
 2)y = 1 and Dx
+ (2D
+ (D
=t
2 and 3x
x = cr e
+ (D
 4)y
5
+ .
2
8
3
+ C2e5t  t
 5
25
and
y = (D  2)x  t
39. From (D  2)x  y
+ 2 = cle
5t
16
11
+ 3C2e + 25 + 25 t.
= _et and 3x + (D  4)y = 7et we obtain (D  l)(D  5)x = 4et so that
Then
192
= sin2t
(D+2)x+(D+l)y
and 5x+(D+3)y
= cos2t
Then
y
and
Cl
cos t
1
. t  3'
2 cos 2t + 3'
7.sm 2t
+ C2 sin
1
193
we obtain (D2+5)y
= 2cos2t7sin2t.
Equations
Exercises 5.1
_
+ 16x = O we obtain
1. From ~x"
x = ct cos s)2 t
+ C2 sin s)2 t
+ kx
= O we obtain
x = ct cos ~
3. Frorn ~x"
+ 72x = O, x(O) =
4. From ~x"
+ 72x
5. Frorn ix"
+ 40x = O, x(O) =
(a) x(1f/12)
(b) x'
= 1/1f vibrations/second.
= ~cos St.
= lOcost.
+ 16x =
O, x(O)
= 1,
and x'(O)
x = cos4t
= 2
we obtain
J5
194
 3.6).
Exercises 5. 1
The period is 7f/2 seconds and the amplitude is .)5/2 feet. In 47f seconds it will make 8 complete
vbrations.
9. From
;tX" + x
+ 40x
= O, x(O)
= 1/3,
and x'(O)
x =  ~ cos 5t
3
If x = 5/24 then t
+ 200x
(a) x =
Vl3.sm(2t + 0.588).
3.
we obtain
+ ~sin 5t = ~
4
12
sin(5t
+ 0.927).
for n
= 0,1,2, ....
= O, x(O)
 0.927).
(e) If x' =
235
cos(10t  0.927)
n = O, 1, 2, '"
= (2n
= O then
If x = 5/12 and
+ 9x
t = fo(0.927
=  cos 3t 
+ 47f/3).
If x'
= V3 we obtain
V3 sin
. 3t
""3
= 3 then t
2.
= V3 sin 37 +
= 77f /18
47f)
3
+ 2n7f/3
and t = 7f /2
+ 2n7f/3
for
= O, 1, 2, ....
+ 192x
14. Let m denote the mass in slugs of the first weight. Let k1 and k2 be the spring constants and
k = 4k1k2/(k1
+ k2)
the efIective spring constant of the system. Now, the numerical value of the
195
Exercises 5. 1
first weight is W
= mg = 32m, so
32m
= klG)
and
32m
= k2G)
From these equations we find 2k1 = 3k2. The given period of the combined system is 27r/w
so w
= 30.
.s:
1/4
= 4k
or
= 7r/15,
= k/m
k = 225.
Solving the second equation for k1 and substituting in the first equation, we obtain
4(3k2/2)k2
3k2/2 + k2
12k = 12k2
5k2
5
= 225.
Thus, k2 = 375/4 and k1 = 1125/8. Finally, the value of the first weight is
W =.32m = k1 = 1125/8
3
3
15. For large values of t the differential equation is approximated by x" = O. The solution of this
equation is the linear function x = clt
+ C2.
decayed to the point where the spring is incapable of returning the mass, and the spring will simply
keep on stretching.
16. As t beco mes larger the spring constant increases; that is, the spring is stiffening. It would seem
that the oscllations would become periodic and the spring would oscllate more rapidly. It is likely
that the amplitudes of the oscillations would decrease as t increases.
17. (a) above
iX" + x' + 2x = O, x(O) = 1, and X'(O) = 8 we obtain x = 4te4t  e4t and x' =
8e4t  16te4t. If x = O then t = 1/4 second. If x' = O then t = 1/2 second and the extreme
displacement is x = e2 feet.
21. From
196
Exercises 5. 1
feet.
23.
(a) From
24.
+ ie8t.
= 1 meter.
~e8t (5  2t) = O when t = ~In a ::::: 0.153 seeond; if x' = te8t (t 10)
t = ~In 10 :::::0.384 seeond and the extreme displacernent is x = 0.232 meter.
(b) x =
25.
(a) From
(b) x
O.lx" + O.4xl + 2x
= e2t
V; [ jg
eos4t  ~
(a) From
(b) x
V;
(~
eos4t + ~
= e2t ~
= O then
sin4t) = ~e2t
= 1 we obtain x
sin (4t +
).
(e) If x = O then 4t+7r /4 = 7r,27r,37r, ... so that the times heading downward are t = (7 +8n)7r /16
for n=O, 1, 2, ....
(d)
1
\
\
\
.5
,,
_"'2
.5
/
/
1
27. From
156 x"
+ 3xl+ 5x = Owe find that the roots of the auxiliary equation are m =  ~13 h/432  25 .
197
Exercises 5. 1
28. From 0.75x"
m=
+ (3x! + 6x = Oand
 ~(3 ~
(3
18 and
(32 
+ ~x' + 6x
C1 = O and C2 =
= 10cos3t, x(O)
Xc
and xp =
13 ( cos 3t
> 3vZ we find that the roots of the auxiliary equation are
+ sin 3t)
18t
+ C2 sinh ~J(32
3/
(32 
18.
(V47
V47 t )
+ C2 sin 2
.
+ sm3t).
Xc
and xp
x = et cos 2t
x
(b)
18t] .
+ 3 sin 2t.
x
(e)
steadystate
3 transJ.ent
.
3
31. From x" + 8x' + 16x = 8 sin 4t, x(O) = O, and x' (O)
xp =  cos 4t so that the equation of motion is
= O we obtain
Xc
C1 e4t
1
x = _e4t
32.
+ te4t
As t
> 00
the displacement x
>
O.
and
1
  cos4t.
+ c2te4t
Xc
1
21 cos 4t + 49 sin 4t + 2e2t
198
C1 cos 4t
+ C2 sin 4t
and
Exercises 5. 1
34. Since x = ~ sin(4t  0.219) 
~e2t
35. (a) By Hooke's law the external force is F(t) = kh(t) so that mx" + f3x' + kx = kh(t).
(b)
e2t(
2 (
t
56
72)
13 cos 2t  13 sin 2t
56
32 .
+ 13 cos t + 13 SIn t.
Xc
= el cos4t + e2sin4t
l sin 8t so that
x
(b) If x
Xc
(e) If x' = ~ cos 4t ~ cos 8t = ~(lcos 4t)(1 +2 cos 4t) = Othen t = 1f /3+mr /2 and t = 1f /6+n1f /2
for n = O, 1, 2, ... at the extreme values. Note: There are many other values of t for which
x' = O.
(d ) x(1f/6 + n1f/2) = V3/2 cm. and x(1f/3 + n1f/2) = V3/2cm.
(e)
x
1
Xc
SIn
38. From x" +9x = 5 sin3t, x(O) = 2, and x'(O) = Owe obtain
2t + t
5 cos 2t.
Xc
= el cos3t+e2 sin3t,
xp = ~tcos3t,
and
5.
(w2
= O we obtain
",(2)so that
x =
Fo
w "'(
2 cos wt +
199
Fo
w "'(
2 cos "'(t.
Xc
Exercises 5. 1
(b)
lim
"(+w
Fo
2
2(cos 1t  coswt) = lim
w  1
"(+w
= (Fot/2w)sinwt
21
Fo
= t sinwt.
2w
= (Fot/2w)sinwt
so that x
and
follows.
(b) If 10 = 1(1  w) then 1
;::::o
42. See the article "Distinguished Oscillations of a Forced Harmonic Oscillator" by T.C. Procter in The
College Mathematics Journal, March, 1995. In this article the author illustrates that for Fo = 1,
>. = 0.01, 1
22/9, and w
1
where A
Fo(w2  12).
xp(t) = (w212)2+4>.212
sm1t+
Now
Fo( 2>'1)
(W212)2+4>.212
where
Fo( 2>'1)
and
200
.
COS1t= Asm(1t+B),
Exercises 5. 1
= O then v
+ 2).2 
(b) If g'(I)
(,2
W2)
,1 = vw
and the resonance curve grows without bound at ,1 =
(c) We identify w2
= k/m =
4, ).
{3/2, and
2  2).2
= J4 
{32/2. As {3; O,
,1 ;
resonance.
(3
yI
2.00
1. 00
0.75
0.50
0.25
1. 41
1. 87
1. 93
1. 97
1. 99
0.58
1. 03
1. 36
2.02
4.01
,t
'1 = w /2.
1(1 cos2,t).
,t
sin3
= sin ,t sin2 ,t
Now
sin(A
+ B) + sin(A
 B) = 2sinAcosB
so
sin ,t cos 2,t = ~[sin 3,t  sin ,t]
and
sin3
,t
'4 sin ,t
Thus
x"
+ w 23.
x = '4 sm v t  'l'3t
4 sm
,.
,3
,1
201
= w, and when
Exercises
(e)
5. 1
yl=1/2
yl=l
y2=1I3
10
n=3
10
5
10
10
+ 2q' + 100q = O we obtain q(t) = e20t(cr cos 40t + C2 sin 40t). The initial conditions
q(O) = 5 and q'(O) = O imply cr = 5 and C2 = 5/2. Thus
q(t)
= e20t (5cos40t+
~sin40t)
:::::)25
+ 25/4e20tsin(40t
+ 1.1071)
and q(O.Ol) :::::4.5676 coulombs. The charge is zero Ior the first time when 40t
+ 0.4636 =
7f
or
46. Solving
and q'(O)
47.
conditions
= C2 = 10. Thus
q(t)=1010e3t(cos3t+sin3t)
and
i(t)=60e3tsin3t.
= 7f /3
+ 100q' + 2500q = 30 we obtain q(t) = ce50t + c2te50t + 0.012. The initial conditions
q(O) = O and q'(O) = 2 imply C = 0.012 and C2 = 1.4. Thus
q(t) = 0.012e50t
Solving i(t) = Owe see that the maximum charge occurs when t
= 2e50t
=
 70te50t
+ 2q' + 4q = Owe obtain Yc = et (cos v'3 t + sin v'3 t). The steadystate charge has the
form Yp = A cos t + B sin t. Substituting into the differential equation we find
(3A
Thus, A
150
100
13cos t + 13 sin t
202
Exercises 5. 1
50. From
Ea
ip(t) = Z
and Z
= JX2 + R2
A = jE'6R2
E'6X2 = Ea. /R2 X2 = Ea
\ Z4 + Z4
Z2 V
+
Z .
51. Thc differential equation is ~q" + 20q' + 1000q
Z = ,j X2 + R2 =
,j X2 + 400 ~
24.0370,
and
Eo = 100 ~ 4 1603.
+ cp)
ip(t) ~ 4.1603(60t
where sin
angle. Now
X/Z
X/R
cp ~ 0.5880 and
ip(t) ~ 4.160~(60t  0.5880).
52. Solving
+ C2 sin 40t).
40t. Substituting
equation we find
(1600A
+ (400C
= 200 sin 60t
+ (2400A
+ (1600C + 400D)
cos40t
1/26,
3/52,
e = 4/17,
and D
state charge is
qp(t)
1
3
26 sin 60t  52 cos 60t
=
4
17 sin 40t
1
17 cos 40t
ip(t) = 
45
sin60t
13
+
203
160
40
cos40t   sin40t.
17
17
+
Exercses 5. 1
53. Solving ~q"
+ 10q' + 100q =
The initial
Thus
As t
+ 00,
q(t)
+
3/2.
L"( 
l/C"(.
+ R2
and
where Z = JX2
= 1/L,,(2.
+ 10q =
Asin"(t + Bcos"(t.
c cos
= 100/(100
= q'(O) = O imply
q(t) =
 ,,(2)and B
Cl = O and C2=
= O. Thus, qp(t) =
102(lOsin"(t"(sin10t)
100  "(
i(t) =
100"(
2(cos"(tcoslOt).
100  "(
L dt2 + C q = E(t).
Then q(t) = Clcos (t/..JLC) +C2 sin (t/..JLC)
qp(t) into the differential equation we find
(~
1
1
EoC
q(t) = Cl COS ..JLC t + C2sin ..JLC t + 1 _ LC"(2 cos vr.
204
Exercises
The initial conditions q(O) = qo and q'(O) = io imply
The current is
Cl
= qo  EoC/(l
.
Cl.
1
C2
1
EoC,.
2(t) =  ..JLC sin ..JLC t + ..JLC cos ..JLC t  1 _ LC,2
EOC).
qO 1_ LC,2
SIn
LC,2) and
t  1_
58. When the circuit is in resonance the form of qp(t) is qp(t) = At cos kt
= io..JLC.
,t
EoC,.
LC,2
1
SIn..;re
C2
5.2
+ Bt
SIn
,t.
= O and B
= Eo/2kL.
The charge is
= qo and
C2
Exercises 5.2
C2 =
Wo
+C4X + 24EIX .
=O
WO
6C4+ EIL = O.
Solving, we obtain
C3
= woL2/4EI
and
C4
= woL/6EI.
y(x) = ~(6L2x2
24EI
205
The defiection is
 4Lx3 + x4).
Exercises 5.2
(b)
~~x
20.40.60.81
1
+C4X
Wo
+ 24EIX
= O and
C3
= O. .The conditions at x
C2L
C2
Wo
Wo
+ C4L + 24EIL
6C4L
Solving, we obtain
= L give the
= WOL3/24EI
and
C4
+ 2EIL
system
=O
= O.
= woL/12EI.
y(x) = 2:;I(L3x
 2Lx3
The defiection is
+ x4).
(b)
+~x
0.2 0.4 0.6 0.8
1 y
y (x ) = C + C2X + C3X
Wo
+ C4X + 24EIX
+ C4L3 + 2:;IL4
Wo
2C3
+ 6C4L + 2EIL
206
=O
2
= O.
Exercises 5.2
Solving, we obtain
C3
= woL2/16EJ
and
C4
= 5woL/48EJ.
y(x) = ~(3L2x2
48EJ
 5Lx3
The defiection is
+ 2x4).
(b)
Cl
el
o, y'(O)
1r
sm LX'
wOL4.
C3L
Solving, we obtain
C3
wO
+ C4L + El7r3L
= 3woL2/2El7r3
woL
y(x) = 2El7r3
= O,y(L)
(22L
and
X
=O
2e3
+ 6C4L = O.
C4
= woL/2El7r3.
+ 3Lx
The defiection is
2  X 3 + ;2L3
1r)
sin LX
(b)
1 Y
(e) Using a CAS we find the maximum defiection to be 0.270806 when
5. (a) The general solution is
207
= 0.572536.
Exercises 5.2
The boundary conditions are y(O) = o, y"(O) = O, y(L)
give C1= Oand
C3
= O,y"(L) = O. The
Wo
6C4L + 6EIL
Solving, we obtain
C2
= 7woL4/360EI
and
C4
= O.
= woL2/36EI.
The defiection is
(e) Using a CAS we find the maximum defiection to be 0.234799 when x = 0.51933.
6.
(a) Ymax
wOL4
= y(L) = 8EI
5woL4
(e) Ymax = y(L/2) = 384EI
(d) The maximum defiection in Example 1 is y(L/2) = (wo/24EI)L4/16
1/5 of the maximum displacement of the beam in Problem 2.
WOL4/384EI, which is
{P
(P Wo 2 woEI
El x + C2 sinh V El x + 2P x + ;r .
y = C1cosh V
Setting y(O) = O we obtain
Cl
woEI
{P
. (P wo 2 woEI
y = ;r cosh V El x + C2 sinh V El x + 2P x + ;r'
Setting y' (L) = O we find
C2
(fu
w~~I sinh
fu
L  W;L) /
208
fu fu
cosh
L.
Exercises 5.2
8. The general solution of the differential equation is
y
/P
. /P
Wo
woEI
/P
. /P
El x + cz sin V El x +
'WoEI
y =  :>2 cos V
Setting y' (L)
:s: O the
'WoEI
= O we find
e2 =
9. For A
Wo
2P x + p2 .
/P woEI. /P WOL)
( V El p2 Slll V El L  P
V/P
El cos V/P
El L.
problem is y = O. For A
> O we have
gives
V>n=nn
or
A=n2,
n=1,2,3,
....
= O implies el = O, so
y
() = e2sin V> = O
gives
,n
VA=nn
or
A=16n,
n=1,2,3,
....
The eigenvalues 16n2 correspond to the eigenfunctions sin 4nx for n = 1, 2, 3, ....
11. For A
:s: O the
Now
209
Exercises 5.2
gives
r:
VA
L =
(2n  1)7f
2
or
A=
(2n  1)27f2
' n = 1, 2, 3, ....
4L2
.
2
.
(2n  1)7f
The eigenvalues (2n  1) 7f2/4L2 correspond to the eigenfunctions cos
L
x for
n
= 1,2,3, '" .
Now y(O)
= O implies CI = O, so
y' (~)
= e2,f).. cos,f)..
=O
gives
,f).. ~ = (2n;
1)7f or
< O the only solution of the boundaryvalue problem is y = O. For A = O we have y = CIX +cz.
= CI and y'(O) = O implies el = O. Then y = e2 and y'(7f) = O. Thus, A = O is an eigenvalue
with corresponding eigenfunction y = 1.
13. For A
Now y'
Now
y' (x) = el ,f)..sin ,f)..x
and y'(O)
= O implies e2 = O, so
gives
J>: 7f = n7f
or
A = n2, n
= 1, 2,
3, .. , .
= O, 1, 2, ....
14. For A :::;O the only solution of the boundaryvalue problem is y = O. For A > O we have
y
210
(1)
Exercises
5.2
sin ~ tt I
.
1\
= 2 sin ~
sin V A tt
1\
tt COS V
1\
A tt = sin 2v A tt = O.
Then
n2
A = 4;
or
n = 1, 2, 3, ....
When n = 2k  1 is odd, the eigenvalues are (2k  1)2/4. Since cos(2k  1)7f/2 = O and
sin(2k  1)7f/2 =1 O, we see from either equation in (1) that C2 = O. Thus, the eigenfunctions
corresponding to the eigenvalues (2k  1f /4 are y = cos(2k  l)x /2 for k = 1, 2, 3, ....
Similarly,
when n = 2k is even, the eigenvalues are k2 with corresponding eigenfunctions y = sin kx for
k
= 1, 2, 3, ....
m=~(2J44(A+1))
For A
=1!=1.
< O we have
y = e x
(q cosh R x + C2sinh R x) .
y(5) =
ci
sinh
C2e5
5R =
> O we have
y = e x
(C1
COS ~
or
A =
X + C2sin ~
x) .
5~
The eigenvalues n27f2/25
16. For A
= nit
:25'
n = 1,2,3, ....
n7f
correspond to the eigenfunctions ex sin 5x for n
y = O. For A = 1 we have
y = CIX+ C2. Now y' = C1and y'(O) = O implies C1= O. Then y = C2and y'(l)
is an eigenvalue with corresponding eigenfunction y
211
1.
1, 2, 3, ....
= O. Thus, A = 1
Exercises 5.2
For A
> 1 we.have
y = ClCOS~x+c2sin~x.
Now
y' = cl~sin~x+c2Jr+lcosJr+lx
and y'(O) = O implies C2= O, so
y'(l)
= clVI+!sinJr+l
gives
~
= n1f
or
/\ = n21f2 1,
n = 1,2,3, ....
The eigenvalues n21f2  1 correspond to the eigenfunctions cos titcx for n = O, 1, 2, ....
17. For A = O the only solution of the boundaryvalue problem is y = O. For A
=f. O we have
=O
implies Cl = O, so
y(L) = c2sinAL
gives
AL = tit: or
A=
=O
n1f
L'
ti
= 1,2,3, ....
n1f
LX
for
ti
= 1, 2, 3, ....
problem is y = O. For A
=f. O we have
or
A=,
2n  1
n=l,2,3,
....
1x for n = 1, 2, 3, ....
= q cos(~
lnx) + c2sin(~
lnx).
lnx).
= O implies
C2Sin~1f
212
=O
Exercises 5.2
or v0."'Jr= n'Jr, n = 1, 2, 3, ....
A = n2,
= 1, 2, 3,
...
and
= sin (n In x).
problem is y = O.
To obtain the selfadjoint form we note that the integrating factor is (1/x2)ef
dx]
l/x.
That is,
r
)1
20.
e"
1
sin(nlnx)sin(mlnx)dx
x
m =J n.
C2e
O implies
= O,
= C1XvY + C2XvY.
and Cl = O, so that
y = c2sin(v0."lnx).
Now y'
c2(v0."/x)cos(v0."lnx),
so y'(el)
1, 2, 3, ....
= c2ev0."cosv0." = O implies
2n  1
y = sin ( 2'Jrlnx
21. For A = O the general solution is y = el +C2Inx.
= O we obtain Cl = O and
= C2 = O and y
el.
Since y'(e2) = O for any Cl we see that y(x) = 1 is an eigenfunction corresponding to the eigenvalue
A = O.
For A < O, y
~.
Y = clsm(VA
and y'(l)
= e2v0." = O implies C2 = O.
r:
v0."
r:
lnx) +C2COS(VA lnx),
x
Finally,
y'(e2)
= (elv0."/e2)sin(2v0.")
22.
C1X
1/'">
cos
(/4A 2  1)lnx
213
(/2
1 lnx. )
O implies
Exercises 5.2
Since In 1 = O, the boundary condition Y(l) = Oimplies cr= O. Therefore
y = C2Xl/2sin (~
lnx).
(n27r2 + 1) /4,
=0
...
and
= xl/2sin
(n27rlnx).
problem is y = O.
O we obtain Cl
+ C2xl/2In
x.
O, so y
= (1/x2).
(1/x2)eJ(2/x)dx
x2 = 1.
[x2y'] + Ay
= O.
Identifying the weight function p(x) = 1 we can then write the orthogonality relation
le
or
i .xl/2 sin
(~7r
In
lnx) sin
lnx)
(n27rlnx) dx = O,
criticalload
dx = O,
m
m i=
n,
i= n.
will be P4.
= Cl COS
ffi
+ C2sin
214
ffi
+ O.
Exercises 5.2
q =
cos
:;
L = O.
then
I"'if
y = CICOS
From y(O)
+ C2sin
I"'if
Wn
= mrVT /
LvP,
n = 1, 2, 3, ....
n1l'
y (x) = C2sin T x,
x.
L = mr, n = 1, 2, 3, ....
n = 1, 2, 3, .. . ,
where C2=1 O.
26. (a) When T(x) = x2 the given differential equation is the CauchyEuler equation
+ 2xy' + pw2y = O.
x2y"
+ 2m + pw2 = m2 + m + pw2 = O
are
mI =
/
"21  "21y4pw
 h,
+ C2XI/2sin(A In x)
215
Thus,
Exercises 5.2
An
y(e)
The condition
= n7r, n = 1, 2,3,
O requires
....
O. We obtain
a nontrivial
solution
when
But
An
Solving for
c2el/2sinA
= J4pw~
 1j2
= n7r.
gives
Wn
The corresponding
solutions
are
1
n=l
n=3
x
eX
1
1
n=2
conditions
1
equation
is m2
+m =
= Ul
1
+ 1) =
m(m
O so that
+C2
_ (UO  Ul) b
b
a
 a
Cl 
an
u(r)
= uo,
+ C2.
clr1
The boundary
= Ul.
clbl +C2
Solving gives
_ ulb  uoa
b
.
 a
C2 
Thus
UO  Ul) ab
ulb  uoa
ur() = (
+
ba
r
ba
28.
The auxiliary
equation
is m2 = O so that u(r)
+ C21n a
+ C21n r. The
Cl + C21n b = Ul.
= Cl
= uo,
 uolnb
ln(ajb)
ullna
C =
Thus
+ Uo 
and
.
boundary
conditions
u(a) ~ uo
Solving gives
Uo  Ul
C2= ln(ajb)'
Ul 1
_ Uo ln(rjb)  ulln(rja)
nr ln(ajb)
ln(ajb)
29. (a) The general solution ofthe differential equation is y = Cl cOS4X+C2 sin4x.
From yo = y(O) = Cl
+ c2sin4x.
we see that y
satisfy yo
= yocos4x
boundaryvalue
From Yl
Whenever
(e) When yo
=f. Yl there
will be no solutions.
216
is a solution of the
Exercises 5.2
(d) The boundaryvalue problem will have the trivial solution when YO
will not be unique.
30.
(a)
Yl
O. This solution
Y = cos 4x
will be a unique solution when sin 4L
(1.sm4CO~4L) sin 4x
(b) There will be infinitely many solutions when sin 4L = O and 1  cos 4L
= br /4 where
k = 1, 3, 5, ....
(d) There can be no trivial solution since it would fail to satisfy the boundary conditions.
31. (a) A solution curve has the sarne ycoordinate at both ends of the interval
[7r,
7r
For A
or
=O
2Cl7r
= O is Y = ClX + C2.
= O. Thus,
Cl
value problem.
For A
< O we have Y
= Cl cosh AX
y(
7r)
+ C2 sinh AX.
= Cl cosh( An)
= Cl cosh A7r 
+ C2 sinh( C2
sinh A7r
or
2C2
Cl
C2
= O. Thus, for A
A7r)
+ C2 sinh A7r
The second boundary condition implies in
is Y = O.
For A
> O we have Y
= Cl
COS AX
y(
7r)
+ C2 sin AX.
=
Cl
cos(  A7r)
= Cl COS A7r
= Y(7r)
= Cl
+ C2 sin( C2
sin A7r
COS A7r
217
An)
+ C2 sin A7r
Exercises 5.2
or 2C2 sin A71" = O. Similarly, the second boundary condition implies 2eA sin A71"
C2
C2
C2
= O. If e =
=  sin nx,
we may assume without loss of generality that the eigenvalues are An = n, for n a positive
integer. The corresponding eigenfunctions are Yn = cos nx and Yn
= sin nx.
(e)
3
y = 2 sin 3x
y = sin 4x  2 cos 3x
> O the general solution is y = Cl COS V).. X + C2 sin V).. x. Setting y(O)
so that y = C2 sin V).. x. The boundary condition y(l) + y'(l) = O implies
C2 sin
roots.
C2
Since An =
< O. For A
general solution y
Cl
= O,
(b) We see from the graph that tan x = x has infinitely many
Xn
Cl
Taking
An
= O we find
C2
= O, so y
x;;,
tan x
5
= O has
(e) Using a CAS we find that the first four nonnegative roots of tan x = x are approximately
2.02876,4.91318,7.97867, and 11.0855. The corresponding eigenvalues are 4.11586,24.1393,
63.6591, and 122.889, with eigenfunctions sin(2.02876x), sin( 4.91318x), sin(7.97867x), and
sin(11.0855x).
218
Exercises 5.3
Exercises 5.3
1. The period corresponding to x(O)
_
= 1 is approxi
= 1, x'(O)
2
x
10
8
6
4
2.___ +t
2
= 1 is approx
10
8
a periodic solution.
6
4
2
219
Exercises 5.3
5. Frorn the graph we see that [xII ~ 1.2.
.:c
3
2
1
1
7. Since
xeO.01x
for small values of
X,
a linearization is d2X2
.:c
8.
+X
J ~ x
= O.
3
For x(O)
= 1 and x'(O) = 1 the oscillations are symmetric about the line x = O with amplitude
vI2 and
x'(O) = 1 the oscillations are symmetric about the line x = Owith amplitude a
220
Exercises 5.3
little greater than 2.
For x(O) = 2 and x' (O)
2 with small
amplitude.
For x(O) = 2 and
Xl
For x(O) = .;2 and XI(O) = 1 the oscillations are symmetric about the line x = Owith amplitude
a little greater than 2.
9. This is a damped hard spring, so al! solutions should be oscillatory with x ) O as t ) oo.
x
5
4
3
2
Jo
2
221
Exercises
11. .
5.3
k l,
0.01
15
k1
3
15
x
kl
20
k1
100
2
2
3
3
When kl is very small the effect of the nonlinearity is greatly diminished, and the system is close
to pure resonance.
12. (a)
= 0.000471
0.000472
. :::;0.000472.
(b)
0.078
= 0.079
The system appears to be oscillatory for 0.077 :::;k1 < O and nonoscillatory for k1 :::;0.078.
222
Exercses 5.3
13. For,\2 
W2
> O we choose ,\
2 and w
1 with
x(O)
systems.
3
= (dv/dy)(dy/dt)
= vdv/dt,
so vdv/dy
= _gR2/y2.
But, by the
Separating variables
2dy
y2
and
gR2
y
=+c.
v2 = 2g(b) As y ;
1 2
v
00
 2gR+v6.
= 2gR
and va
= V29R.
VD
= V2(32)(4000)(5280) ::::::
36765.2 ft/s::::::25067 mi/hr.
15. (a) Intuitively, one might expect that only half of a 10pound chain could be lifted by a 5pound
force.
(b) Since x = O when t
~j5fi ::::::
2.3717.
Since the
x
10
8
x'(t)
x(tm)
0.5
1.5
2.5t
of the momentum generated by the force. When the chain is 5 feet high it still has a positive
velocity of about 7.3 ft/s, which keeps it going higher for a while.
16. (a) Setting dxf d:
= vdv/dx,
so (L  x)vdv/dx
223.
 v2
Exercises
5.3
v
so VV2
+ Lg
L dV=Ldx
9
x
and
21n(v2+Lg)=ln(Lx)+lnc,
= O,V =
O, and c
Ly'Eg.
we get
dx
dt
= V(X)
V Lg(2Lx
 X2)
L  X
.
= dt
=L
x(t)
= O and
V 12 
y'2Lx  x2
y'Eg
=t+cr
and
y'2Lx  x2/y'Eg
= t.
Lgt2
= dx = =VL=L=g=t=
and
v(t)
dt
/L  gt2
;;.
(e) The predicted velocity of the upper end of the chain when it hits the ground is infinity.
17. (a) The weight of x feet of the chain is 2x, so the corresponding mass is m
= 2x/32 = x/16.
The
only force acting on the chain is the weight of the portion of the chain hanging over the edge
of the platform. Thus, by Newton's second law,
!!:_(mv)
dt
and x dv/dt
xvdv/dx
= !!:_ (.3:_v) =
+ v2 =
dt
16
32x.
]_(x
16
dv
dt
+ V dX) = ]_ (x
dt
16
dv
dt
+ v2)
= 2x
(dv/dx) (dx/dt)
v dv/dx,
so
dv
= O. This
is
+ v2 = 32x.
(b) We separate variables and write the differential equation as (v2  32x) dx
+ xv
+ x2vdv =
3i x3 = c. Letting
3 and v
v we get
dx
8y'x3  27
=v =
dt
vf3x
3 ::; x ::;8.
x
y' 3
x  27
dx =
8
M dt
y3
and
(X
13
224
~x2v2  2:fx3, so an
s
ds = ~ t + c.
y's3  27
vf3
Exercses 5.3
= O and
t = J3 (X
s
8 13 vs3 
ds.
27
We want to find t when x = 7. Using a CAS we find t(7) = 0.576 seconds.
18. (a) There are two forces acting on the chain as it falls from the platform. One is the force due to
gravity on the portion of the chain hanging over the edge of the platform.
This is FI = 2x.
The second is due to the motion of the portion of the chain stretched out on the platform. By
Newtori's second law this is
Po = _r!_[
]
2
dt mv
= _r!_[(8X)2
dt
32
8  x dv
1 dx
dt  16v dt
=~
] = _r!_[~v]
dt
16
1 [
dv
2]
= 16 (8  x) dt  v .
d
From dt [mv] = FI  F2 we have
_r!_ [2X v]
dt 32
dv
x+v
dt
16
=32x(8x)+v
dv
x dt
= 32x  8  + x 
dv
dt
dv
8 dt
= 32x.
dv
dt
(b ) Integrating v dv
= 2x _ 2_
(dv/dx)(dx/dt)
= v dv/dx, so
dv
dv
dv
8 dt = 8v dx = 32x and v dx
= 4x.
v2 = 4x2  36 and v =
V4x2
= 3, we have e
= 18. Then
obtain
dx
v'X2.=9
= 2 dt
and
x
coshI'3
= 2t + q.
= 1 and
(e) To find when the back end of the chain will leave the platform we solve x (t) = 3 cosh 2t =
8. This gives tI = ~coshl
6sinh(cosh1
~)
= 2V55
~ ~
0.8184 seconds.
~ 14.83 ft/s.
225
Exercises 5.3
(d) Replacing 8 with L and 32 with g in part (a) we have Ldv/dt
dv
L dt
.12
Integrating we get:zv
for v we find
dv
= Lv dx = gx
dv
v dx
and
= gx. Then
= LX.
g
v(x)=
/!!'_x2!!'_x5.
L
L
Then the velocity at which the end of the chan leaves the edge of the platform s
v(L)
j f(p  x5)
19. Let (x, y) be the coordinates of 82 on the curve C. The slope at (x, y) is then
dy/dx
= (Vlt
 y)/(O  x) = (y  vlt)/X
or
xy'  y
+ y"
dt
y = vI dx
xy
"
dt ds
ds dx
= Vl
JI + (y')2)
= rJ1 + (y')2.
+ u2
'!._dx
x
snh 1 u = r In x
+ In e =
ln(cz")
u = snhln cz"]
!(ex
At t = O, dy/dx
If r > 1 or r
1_) .
dy =
T __
dx
2
cz"
O and x = a, so O = ca" l/eaT. Thus e = l/aT and
(X)l+T
a
  1
1r
226
(X)lT]
a
+ el
= vlt.
Exercses 5.3
When t =
o, y
+ r) 
= O and x = a, so 0= (a/2)[1/(1
1/(1  r)l
+ CI.
(~)l+r
__1 1 r (~)lr]
+ ~.
a
a
1  r2
y = 9:_[_1
2 1 +r
SI.
Next. if r
> 1, then
< 1, then
VI
VI
< V2 and y
00
as x _, 0+.
= ar/(l
r2) when
x = O. In other words, when the submarine's speed is greater than the ship's, their paths will
interseet at the point (O,ar/(l
r2)).
y = ~ [x2 _ ~ In x] + C2.
2 2a a
When t
and
= O, y = Oand x = a, so O = (1/2)[a/2
y=~[X2
2 2a
Sinee
u : 00 as x
~lnx]
 (l/a) In al + C2 Thus
= (1/2)[a/2
=~[~(x2a2)+~ln~].
2 2a
a
~[9:_~lna]
2 2 a
C2
 (l/a) In al
SI.
20. (a) Let (r, e) denote the polar eoordinates of the destroyer
SI.
When
SI,
SI
of S2 are (3, e2), where e2 is unknown. In other words, the distanees of the ships from (O,O)
are the same and r(t)
SI
= 15t then gives the radial distanee of both ships. This is neeessary if
is to intereept S2.
Using ds/dt
= 30 and
dr/dt
= 15 then
gives
2
900
de
dt
=
e(t) =
When r
= 3, e = O, so C = y'3
e(t) =
y'3
t
v'3 In t + C = v'3 In ;5 + c.
In(1/5) and
v'3 In:3 .
Exercises 5.3
Thus r = 3eB/v'3, whose graph is a logarithmic spiral.
(c) The time for SI to go from (9, O) to (3, O) =
for sorne angle
i hour.
/3, where O < /3 < 27l'. At the time of interception t2 we have 15t2
= 3e/3/v'3 or
t =
1
1
+ _e/3/v3
< (1 + e27r/v3).
;;
;;
fact that the motion is oscillatory and the velocity (or its square) should be negative when the
spring is contracting.
22. (a) The approximation is accurate to two decimal places for
el = 0.3, and
el = 0.6.
0.8
0.6
0.4
0.2
0.20.40.60.8
(h)
0.3r~
0.2
0.11_
+'0"."""2
"0'.
40
e
o.
O.
O.
o.
o.
o.
O.
0.1
O.
.2
0.4
0.3
1.21.4
O.
t
The thinner curves are solutions of the nonlinear differential equation, while the thicker curves
are solutions of the linear differential equation.
23. (a) Write the differential equation as
d2
dt2 +w2sine
= O,
2
Comparing the apparent periods of the graphs, we see that the pendulum oscillates faster on
the earth than on the moon.
228
Exercises
5.3
d2e
dt2 + w2e = O,
where w2 = g/e. When
and
= 32,
e = 3, e(O)
= 1,
general solution is
2
dt2 + e = O
is e(t)
Setting e(t)
= tanI(1r/4)
::::;:
0.66577.
h (J3 1)/4
and
= h (J3 + 1)/12.
e(t) = _:__ ~t _
12
J2 (J3 
1) t2 +
J2 (J3 + 1) t3 + .. '.
= O we obtain
tI
72
= 1r/4::::;:0.785398.
(d) Setting
_:_~t
J2(J31)t2=0
12 3
8
and using the positive root we obtain tI ::::;:
0.63088.
(e) Setting
_:_ _ ~ t _
12
J2 (J3 
1) t2 +
J2 (J3 + 1) t3 = O
72
229
Exercises 5.3
we find tl ::::::
0.661973 to be the first positive root.
5
0.4
(g) To find the next two positive roots we change the interval used in
0.4
h=O.l
tn
en
tn
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
1.70
1.80
1.90
2.00
0.78540
0.78523
0.78407
0.78092
0.77482
0.76482
0.75004
0.72962
0.70275
0.66872
0.62687
0.57660
0.51744
0.44895
0.37085
0.28289
0.18497
0.07706
0.04076
0.16831
0.30531
1.70
1.71
1.72
1.73
1.74
1.75
1.76
1.77
1. 78
1.79
1. 80
0.07706
0.06572
0.05428
0.04275
0.03111
0.01938
0.00755
0.00438
0.01641
0.02854
0.04076
h=O 001
1. 763
1. 764
1.765
1.766
1.767
1.768
1.769
1.770
0.00398
0.00279
0.00160
0.00040
0.00079
0.00199
0.00318
0.00438
230
en
1. 8 ft., since k = 4.
2. 271'/5,since
~X"
+ 6.25x
= O.
~sin4t.
4. True
5. False; since an external force may existo
6. False
7. overdamped
8. From x(O) = (V2/2)sin<t> = 1/2
= 1/V2,
fourth quadrant. Since x'(t) = V2cos(2t+<t, x'(O) = V2cos<t>= 1 and cos<t>> O. Thus <t>
is in the
fourth quadrant and <t>= 71'/4.
9. The period of a spring mass system is given by T
where k is the
spring constant, W is the weight of the mass attached to the spring, and 9 is the acceleration due
to gravity. Thus, the period of oscillation is T = (271'/..jJg)JW.
is W, then (271'/..jJg)JW
= 3 and
(271'/..jJg)y'W
 8 = 3/2
or W = ~(W  8). Solving for W we find that the weight of the original mass was 14.4 pounds.
10. (a) Solving
iX"
+ 6x
(e) x'(371'/16) = O
(f) If x' = O then 4t + 371'/4= 71'/2+ n71' or t = 371'/16+ n71'.
11. Frorn
~X"
+ ~x' + 2x
= O, x(O)
= 1/3, and
x'(O)
= Owe obtain
e2t  !e4t.
12. From x" + /3x' + 64x = Owe see that oscillatory motion results if /32
~X"
:::::
If x'(t) = O, then
4.094.
= cos"(t + sinyz in the form x" + 634x = Bcos"(t + Bsin"(t we identify .x = O and
w2 = 64/3. The system is in a state of pure resonance when "( = w = j6473 = B//3.
231 .
xp =
t),
d
26
2Sv2
x=e4t ( cos2v'2t+sin2v'2t
17
17
) +_e8 t.
17
18. (a) Let k be the effective spring constant and Xl and X2 the elongation of springs kl and k2. The
restoring forces satisfy klXI = k2X2 so X2
= (kjk2)XI.
k ( k2 ~ kl )
= kl
k = klk2
kl + k2
1
1
1
k = kl + k2 .
(b) From kl = 2W and k2 = 4W we find l/k = 1/2W +1/4W = 3/4W. Then k = 4W/3 = 4m9/3.
The differential equation mx" + kx = Othen becomes x" + (49/3)x = O. The solution is
x(t) = c cos 2~
t + C2sin 2~
t.
The initial conditions x(O) = 1 and X/(O) = 2/3 imply CI = 1 and C2 = 1/.,f39.
(e) To compute the maximum speed of the weight we compute
Xl (t)
= 100sin50t, q(O)
= O, and
(a) q =
1~O sin
(b) i =
(e) q
100t +
= O when sin50t(1
cos50t)
= Oor t = mr/50
for n
= 0,1,2,
....
d2q
dq
1
L dt2 + R dt + e q = E(t).
Using ql(t)
= i(t)
L dt + Ri +
e1 q = E (t).
232
C2sin 100t,
+ Ri(t) + (1/C)q(t)
(b ) Frorn Li'(t)
+R
di
dt
+ ei = E
(t).
E(O) or Li'(O) +
>
[E(O)  b qo  Rio] .
Cl COS
ci
Now y(O)
Cl
and y(21r)
= y(21r) implies
ci
1, 2, 3, ... , with corresponding eigenfunctions cos nx and sin nx. When X = O, the general solution
is y = ClX + C2 and the corresponding eigenfunction is y = 1.
For .A
y(21r) =
Cl
Cl
cosh H x
+ C2 sinh 21rH,
cosh 21rH
so y(O)
+ c2ewt.
Cl
= O. Thus,
and
there
< O.
+ C2 sinh H x.
= y(21r)
= gsinwt.
+ Bcoswt.
Substituting
and "
is r(t)
g/2w 
WCl
+ WC2 = Vo.
C1 = (2w2ro
Solving for
+ 2wvo
Cl
and
 g)/4w2
C2
we get
and
C2
Cl
+ C2
= ro and
so that
r ()t =
2w2ro
+ 2wvo
4w2
 9 wt 2w2ro  2wvo
e +
4w2
+9
wt
(b) The bead will exhibit simple harmonic motion when the exponential terms are missing. Solving
q = O, C2 = O for ro and vo we find ro = Oand vo = g/2w.
To find the minimum length of rod that will accommodate simple harmonic motion we determine the amplitude of r(t) and double it. Thus L = g/w2.
233
(d)
14
(e) For each vo we want to find the smallest value of t for which r(t) = 20. Whether we look for
r(t) = 20 or r(t) = 20 is determined by looking at: the graphs in part (d). The total times
that the bead stays on the rod is shown in the table below.
Vo
When vo
10
16.1
15
17
20
20
20
20
20
1.55007
2.35494
3.43088
6.11627
4.22339
and substituting into the differential equation, we get 12A sin 4t  12B cos 4t = sin 4t. Thus,
(o:
x(t)
~(o:
+~)
sin2t 2
3
= ~(o: +~)sin2t 2
3
=
~ sin4t
12
~(2sin2tcos2t)
12
tt
O<t<.
 2
For x"
+x =
+1=
O, so
Xc
= el cos t
+ e2 sin t.
Letting
= A sin 4t + B cos 4t and substituting into the differential equation we get 15A sin 4t 15B cos 4t = sin 4t. Thus, A = l5 ' B = O, and xp = l5 sin 4t. The general solution of
xp
234
Cl
= ex+
C2 sin
and
C2
t
!5
= O,
= O. Thus
= cos t [( ex+
D
15
1~ sin t cos 2t ] ,
37r
tt
<t<.
2   2
(b ) The velocity at the start of the second cycle is x' (37r/2) = ex+
(c) Using results from part (a), x =
x(37r/2)
= O and x'(37r/2)
Cl
= ex+
cos 2t +
125'
C2
sin 2t 
which imply
Cl
125 .
112sin 4t.
O and
C2
fs).
Thus
2.)
x (t) =  ~ (ex +
sin 2t  2_ sin 4t
2
15
12
= ~(ex+
175)sin2t
= sin2t[~(ex+
Cl
COSt +
C2
:2(2sin2tcos2t)
2.)
 ~coS2t]
15
6
sin t 
15
'
37r
2 <
 t<
 27r.
Cl
= O and
C2
= (ex +
sin t  2_ sin4t
15
= (ex +~)
sint  2_(2sin2tcos2t)
15
15
=  (ex+ ~) sin t  ~ (2 sin t cos t cos 2t)
15
15
= sin t[ (ex + ~)  2_ cos t cos 2t]
15
15
'
(d)
2
2
ex=l
ex=2
235
..
l~).
Thus
+ e2 sin t
+ Bt
sin t.
conditions imply el = Oand e2 = O,so the solution of the initialvalue problem is x( t) = ~t sin t.
(b) We have
Xc
+X
= sin 4t when
X ::::: O and
x"
+ 4x
= sin 4t
< O. Since the initial velocity is positive, we solve first for x :::::o. The solution of
x" + x = sin 4t is x( t) = el cos t + C2 sin t  t\ sin 4t, as seen in Problem 1(a). The initial
conditions x(O) = Oand x'(O) = 1 imply el = Oand c2 = ~~
, so the solution of the initialvalue
when x
problem is
19
19
15
=  sin t
=
19
4
)
(sint) ( 15  15 costcos2t ,
0S;tS;7L
The first positive value of t for which x(t) is zero is t = 7r, at which time x'(7r) =
so x
x(7r) = O, x'(7r)
i~.The solution
= O is t
7r S; t S; 3; .
x'(37r/2) = ~~ > O, so
> O immediately after t = 37r/2. The next initialvalue problem is then x" + x
x(37r/2) = O, x' (37r/2) = ~~. The solution of this initialvalue problem is
x
17
n < O,
( 17
4
)
15  15 sintcos2t,
37r
2'"
S; t
S;
= sin4t,
57r
2""
12
236
 ~ cos 2t),
6
Chapter
= O. In this case,
5 Related Exercises
115 
which is between 51i/2 and 31i. Thus, for the first two cycles,
(sin t)(
(sin 2t) (
x(t) =
O::;t::;1i
*  i cos 2t),
1i <
t

17
4
(cos t) (15
 15
sin t cos 2t) ,
(sin 2t) (
< 37r2
<
t < 57r
 2
37r
5; < t ::;8.84514.
fs  t cos 2t),
The first part of the next cycle is the solution of the initialvalue problem x" + x = sin 4t,
x(8.84514) = O, x'(8.84514) = 0.28. This solution is approxirnately x(t) = 0.013 cost 0.109 sin t  0.067 sin 4t and is positive on (8.84514,9.42478) with amplitude 0.042.
computations are done with the aid of a CAS having a differential equation solver and graphing
capability.] The solution of x"
+ 64x = sin4t,
1(.
x(O) = O, x'(O) = 1 is
11.)
x(t) = 48 sm4t+2'"sm8t
where
TI
= 0.415458 is found using the rootfinding capability of the CASo The next initial
X(T1)
= O, X'(T1)
T2
TI::;
t::;
T2,
X(T2)
O, X'(T2)
= O, X'(T3)
T3 =
T2::;
t ::; T3,
X(T3)
x(t) = 0.737cos2t
where
T4
 0.217sin2t  0.083sin4t,
T3::;
t::;
T4,
= 4.0453.
< O.
= sin4t,
x(O) = O, x'(O) = 1 is
x(t) = 2_ sin 4t +
20
2.
sin 6t,
15
237
O::; t
< TI,
x 2: O. The solution
TI = 0.571447.
Tl
T2
T3
S;
S;
S;
t S;
t S;
t S;
T2
T3
T4
= 1.24203
= 1.73555
= 2.21355.
1
1
b = 1,
b = 64,
a =4
b = 36,
a =4
a = 25
= 0.5
10
1
1
b = 1,
a =4
= 64,
a =4
= 36,
= 25
= 36,
= 25
1
2
3
1,
a =4
b = 64,
a =4
In both situations, it appears that the amplitude decreases for b = 1 and a = 4, but increases
in the other two cases.
238
Chapter
5 Related Exercses
4. (a) The solutions are obtained numcrically in CAS and plotted below in part (b). Since x'(O) =
1
> O, in each case the first differential equation used is x" + !3x' + 4x
2
x
1
1
2
!3 = 0.01
!3 = 0.5
!3 = 0.1
!3 = 0.5.
239
!3
!3 = 0.1 and
1.
u1m lan+ll
= u1m 12n+lxn+l/(n+1)1
n+ooan
n+oo
2nxn [ti
f:
n=l
(_l)n
n
00 1
I: is the
n=l n
diverges. Thus, the given series converges on [1/2,1/2).
2.
u1m lan+ll
n+ooan
= l'1m !100n+l(x+7)n+l/(n+1)!!
n+oo
100n(x + 7)n/n!
u1m _100 Ix + 7 1= O
n+oo
n + 1
I=
1
(x  5)k+ / lOk+1
1
1
lim
= lim Ix  51= Ix
k+oo (x5)k/10k
k+0010
10
00 (_l)k (10)k
2::
10k
k=l
51 < 1, Ix 
00
2::
1 diverges by the kth term test.
 51
At x
k=l
At x = 5, the
00 (1)k10k
2::
10k
k=l
diverges by the kth term test. Thus, the series converges on (5,15).
lim(k+1)lx11=00,
k+oo
x=ll
6. ex cos x = ( 1  x +
7.
1
cosx
x2
x3
'2
 "6
+
1
2
1~+ir4
,
4
.
fu+'"
6.
x4
24  .,.
x; + ~: _
1
x2
'2
+
;6
x4)
24  '"
x2
5x4
61x6
= 1 + 2 + 41 + 61 + ...
..
240
= 1 x +
x3
x4
3
 "6
+
...
Exercises
Since cos( 7r/2) = cos(
1x
1 3
=   x
2+x
2 4
+ x3
8. 
/2)
T;
 x
16
= 0, the
3
00
2ncnxn1
(7r
7r/2).
00
00
00
n=l
/2,
+ ...
series converges on
6. 1
n=O
+ 6enxn+l
n=2
._,__..._.,
k=nl
00
n=O
~
k=n+l
00
k=l
00
10.
n(n
00
 1)enxn
+ 2 n(n
n=2
00
 1)enxn2
+ 3 nenxn
n=2
n=l
00
00
00
n=4
k=n
00
= 4C2 + (12c3
n=2
k=n
'v'
k=n2
00
00
n=2
n=2
00
11. y' =
00
(_l)n+lxnl,
n=l
y" =
(1t+1(n
 1)xn2
n=2
00
(x
+ 1)Y" + y'
= (x
00
n~
00
00
00
n=2
00
= xo+xo+
 1)xn2
+ (1t+1xnl
n=l
00
(_l)n+l(n_1)xnl+
n=2
k=nl
(_l)n+l(n_1)xn2+
n=3
k=n2
241
00
(_l)n+lxnl
n=2
'v'
k=nl
Exercises
6. 1
00
00
00
+ (_1)k+3(k
= (_1)k+2kxk
k=l
+ 1)xk + (_1)k+2xk
k=l
k=l
00
[(_1)k+2k
_ (_1)k+2k
+ (_1)k+2]
_ (_1)k+2
xk = O
k=l
,_ ~
12. y  ~
n=l
'"
xy
(1)n2n
2nl
22n(n!)2 x
,
" _ ~
y  ~
n=l
(1)n2n(2n
 1) 2n2
22n(n!)2
X
(1)n2n(2n
 1) 2nl
~
22n(n!)2
X
+ ~l
_ ~
+ y + xy  ~l
(l)n2n
2nl
~
22n(n!)2 x
+~
k=n
13. Substituting y
y"  xy
k=n
(_1)n
2n+l
22n(n!)2 x
k=n+l
n(n
00
 1)Cnxn2 
n=2
Cnxn+l
00
n=O
k=n+l
00
+ 1)Ck+2Xk
k=O
'.r'
k=n2
(k + 2)(k
00
= 2C2 + [(k
k=l
Thus
C2= O
(k
+ 2)(k + 1)Ck+2 
Ck+2 = (k
+ 2)(k + 1) Ckl,
Ckl = O
and
1
C3
=
C4
= Cs = O
C6 
1
180
242
k = 1,2,3, ....
Ck_lXk
k= 1
Exercises
and so on. For
= O and
Co
Cl
= 1 we obtain
C3
=O
C5
= C6 = O
C7
= 504
Yl
14. Subst.ituting y =
180
+ ...
+ x2y
Y2 =
1 4
1 7
+ x
+ x
+ ....
12
504
y"
and
.L n(n
00
_ l)cnxn2
n=2
+ .L cnxn+2
00
n=O
k=n+2
00
k=O
'v'
k=n2
.L (k
k=2
00
= 2C2 + 6C3X +
.L [(k + 2)(k
+ I)Ck+2 + Ck_2]Xk
= O.
k=2
Thus
C2
(k
and
C3
=O
Ck+2 = _ (k
Choosing
Co
+ 2)(k + 1) Ck2,
= _
12
CS=C6=C7=0
1
C8 = 672
Co
= O and
Cl
= 1 we obt ain
C4 =
Cs
= _
20
C6=C7=C8=0
cg
1
= 1440
243
=O
k = 2,3,4, ....
6. 1
Exercises
6. 1
1 4
1
= 1  x
+ x
12
672
 ...
Y2 = x  x
.
20
and
1 9
+ x
1440
y"  2xy'
+y
n(n
00
 1)cnxn2
 2
n=2
ncnxn
n=l
~
k=n2
k=n
00
(k
k=O
+ 2)(k + 1)Ck+2Xk 
00
+ cnxn
n=O
k=n
'v"
00
00
k=l
k=O
kCkXk + Ckxk
00
= 2C2 + Co +
[(k + 2)(k
+ 1)Ck+2 
(2k  l)Ck]Xk = O.
k=l
Thus
2C2 + Co = O
(k
and
e
= Co2
2k  1
Ck+2 = (k
Choosing
Co
= 1 and
CI
+ 2)(k + 1) ci,
k=1,2,3,
= Owe find
C2
1
=2
C3
= Cs = C7 = ... = O
1
C4 =
8
7
Co
= O and
Cl
336
= 1 we obtain
C2
= C4 = C6 = ... = O
1
C3 =
6
1
C5
= 24
C7
1
= 112
244
....
Exercises
and so on. Thus, two solutions are
Yl
= 1  x1
1
 x
8
x
7
336
and
+ 2y =
n(n
ncnxn
 1)cnxn2 
n=2
+ 2 cnxn
n=l
k=n
00
kCkXk + 2 Ckxk
+ 2)(k + 1)Ck+2Xk 
= (k
k=O
n=O
k=n
'v'
'v'
k=n2
k=O
k=l
00
+ 2)(k + 1)Ck+2 
(k  2)qJxk
k=l
Thus
2C2 + 2co = O
(k
+ 2)(k + 1)Ck+2 
(k  2)Ck = O
and
C2 =
Co
k2
q+2 = (k
Choosing
Co
1 and
C}
+ 2)(k + 1) q,
k=1,2,3,
....
= O we find
C2 =1
Por
Co
= O and
Cl
C3
C4
=O
C6
= Cs = ClO = ... = O.
Cs
= C7
= ...
=O
= 1 we obtain
C2
C4
= C6 = ... = O
1
6
1
C5=120
C3 =
and
1
Y2 = x  x
6
245
x
1
120
= O.
6. 1
Exercises 6. 1
y = ;~=o Cnxn into the differential equation we have
17. Substituting
00
y"
00
L n(n
+ x2y' + xy =
n=2
+L
ncnxn+l
n=l
"v'
k=n2
k=n+l
00
00
+L
 1)cnxn2
Cnxn+l
n=O
~
k=n+l
00
L (k + 2)(k
+ 1)Ck+2Xk + L (k
k=O
00
 1)Ck_1Xk
k=2
+L
00
Thus
C2
= 06C3 + Co
=O
(k
=O
and
1
C3
= co
6
Ck+2 =  (k
Choosing
Co
+ 2)(k + 1) Ckl,
k = 2,3,4, ....
Co
= O and
Cl
C3
1
=
C4
= C5 = O
C6
=
c3
=O
C4
1
=
C5
= C6 = O
1
45
= 1 we obtain
C7=
252
Yl = 1 x
6
1 6
+x
45
Ck_1Xk
k=l
and
246
1
Y2 = x  x
6
5
232
+x
Exercises 6. 1
18. Substituting y = 2:;;":=0 cnxn into the differential equation we have
= n(n
y" + 2xy' + 2y =
 1)Cnxll2 + 2
= nc.ix"
n=l
n=2
.____,
k=n2
k=n
oo
+2
00
= Cnxn
n=O
k=n
'v'"
oc
(k + 2)(k + 1)Ck+2Xk + 2
kCkXk + 2
Ckxk
k=O
k=l
k=O
oo
= 2C2 + 2co
+ [(k
k=l
Thus
2C2 + 2co
=O
Ck+2 =
Choosing
co
= 1 and
C1
Co

+ 2 Ck,
k = 1,2,3, ....
= Owe find
C2 =1
C3
= Cs = C7 = ... = O
C4
C6
1
=
Co
= O and
C1
"2
= 1 we obtain
= C6 = ... = O
C2
C3
=
C4
2
3
4
e5 =
15
8
c _
c
105
r:
 x
+ ...
247
= O.
Exercises 6.1
19. Substituting y
(x  1)y"
+ y'
00
= L
n(n  1)cnxnl
n=,2
k=nl
00
 L n(n  1)cnxn2
n=2
k=n2
00
+L
ncnxnl
n=l
k=nl
00
= L (k
+ 1)kck+lXk
00
+ 2)(k + 1)q+2Xk + L
 L (k
k=O
k=l
(k
k=O
+ 1)Ck+lXk
00
= 2C2
+ Cl + L[(k + 1)kck+l
 (k
= O.
k=l
Thus
+ Cl
2C2
(k
+ 1)2Ck+l
=O
+ 2)(k + 1)Ck+2 = O
 (k
and
k+1
Ck+2=k+2ck+l'
Choosing
Co
= 1 and
Cl
= Owe find
C2
k=1,2,3,
C2
....
= 2'
C3
1 we obtain
3'
C4
= 4'
Yl = 1
20. Substituting y
Y2
1 2 1 3 1 4
= x + x
+ x + x + ....
2
(x
+ 2)y" + xy'
00
 y = L n(n  1)cnxnl
n=2
k=nl
00
= L(k
+L
00
2n(n  1)cnxn2
n=2
k=n2
00
+L
ncnxn  L Cnxn
n=l
n=O
~~
k=n
k=n
00
00
+ l)kck+lXk + L
2(k
k=O
k=l
+ 2)(k + 1)Ck+2Xk +
00
kCkXk  L Ckxk
k=l
k=O
00
= 4C2 
Co
+L
[(k
 1)cklxk = O.
k=l
Thus
4C2 (k
Co
=O
 1)Ck = O,
1, 2, 3, ...
Exercises
6. 1
and
C2 =
ca
4
(k
+ l)kck+l + (k
ck+2=Choosing
Ca
1 and
Cl
ca
Cl
 l)Ck
2(k+2)(k+1)
k = 1, 2, 3, ....
= Owe find
= O,
= O and
C3
Cl
=
c0480
= O,
C4
24'
1 we obtain
C2 = O
C3
=O
C4
= C5
= C6 = ... = O.
Y2 = ClX.
y"  (x
+ l)y'
 y
L n(n
00
 1)cnxn2 
n=2
n=l
k=n
00
L (k + 2)(k
'v"
k=n2
=
00
L ncnxn
L ncnxnl
n=l
'v'
k=nl
00
+ 1)Ck+2Xk
k=a
00
L Cnxn
n=a
k=n
00
k= 1
k=a
00
L Ckxk
k=a
00
= 2C2 
Cl 
Ca
(k
+ l)Ck+l
k=l
Thus
(k
+ 2)(k + 1)Ck+2 
and
Cl
C2= 2Ck+2 =
Choosing
ca =
1 and
Cl
(k  l)(Ck+l
+ Ck) = O
+ ca
Ck+l + Ck
k + 2 Ck,
= 2,3,4,
....
= O we find
C2 =
2'
C3
6'
249
C4
=6
 (k
+ l)Ck]Xk = O.
Exercses 6. 1
and so on, Por
Co
= O and c = 1 we obtain
C3
2'
C4
1
=4
u: = 1 + 2x + 6x + 6x + ...
and
Y2
= x + 2x + 2x + 4x + ....
+ 1) y"
00
 6y
00
00
 6 L
n=2
k=n2
k=n
00
cnxn
n=O
'v'
k=n
00
00
+ 2)(k + 1)Ck+2Xk 
k=O
00
6L
Ckxk
k=O
6) Ck + (k + 2)(k + 1)Ck+2] xk
k=2
Thus
2C2  6co = O
(k  3)(k
and
C2
= 3co
C3
= Cl
k3
Ck+2 =  k + 1 Ck,
Choosing
Co
= 1 and
Cl
k = 2,3,4, ....
= O we find
C2 = 3
C3
= Cs = C7 = ... = O
C4
=1
C6
1
=
250
= O.
Exercises 6. 1
and so on. For
Co
= O and
ci
1 we obtain
C4
C3 =
C7
C2
C5
= C6 = ...
=O
= O.
Cg
= ...
+x4 
1 6
+ ...
x
5
and
Y2 = X
+ X 3.
(x2
+ 2)yll + 3xy'
00
2:
 y =
00
+ 22:
n(n  l)cnxn
n=2
n=2
k=n
00
2:
00
+ 22:
(k
k=O
k=2
00
nCnxn 
n= 1
~
k=n
k=n2
00
k(k l)CkXk
+ 32:
n(n  1)Cnxn2
(4C2  co)
2:
Ckxk
k=O
k=l
+ (12c3 + 2Cl)X + 2:
[2(k
k=2
Thus
4C2 12c3
2(k
Co =
+ 2c
=O
1)
Ck =
and
1
C2
= Co
C3
Ck+2 =
Choosing
Co
1 and
Cl
1
Cl
6
k2
+ 2k
C2
1
=4
C3
 1
k =
= Owe find
C5
C4 =
C7
= ...
96
251
=O
cnxn
n=O
'v'
k=n
00
kCkXk 
00
2:
2,3,4,....
1) Ck] xk = O.
Exercises
6. 1
Co
= O and
CI =
1 we obtain
C2=C4=C6==0
1
C3 =
6
7
Cs
120
= 1 + 4: x  96 x + ...
1
Y2 = x  x
6
and
7
+x
120
1) y" + xy'
00
 y =
n(n
00
n(n
 l)cnxn 
n=2
+ ncnxn
n=2
n=l
k=n
k(k
~~
k=n2
CXJ
00
 1)cnxn2
00
l)CkXk
cnxn
n=O
k=n
00
 (k
k=O
k=2
k=n
00
k=l
00
Ckxk
k=O
00
[(k
k=2
Thus
2C2 
(k
Co = O
+ 2)(k + 1)Ck+2 + (k
and
ca
= 1 and
CI = Owe
+ I)Ck
C2
= Ca
C3
=O
kl
Ck+2 = k + 2 Ck,
Choosing
 l)(k
= 2,3,4,
find
1
C2 =
C3
= Cs =
C7
= ... = O
C4 =
252
....
=O
Ck] xk
= O.
Exercses
and so on. For co
= Oand C} = 1 we obtain
C2
= C4
C6
= ... = O
C3
= Cs
C7
= ...
= O.
Yl
25. Substituting y =
1 2
1 4
1  x x
and
Y2
= x.
(x  l)y"  xy'
00
+ y = L n(n  1)cnxn1
n=2
k=nl
00
00
00
L(k+1)kck+1Xkk=l
2C2
00
L(k+2)(k+1)Ck+2Xkk=O
+ L cnxn
n=O
k=n
'v"'
00
LkckXk+
k=l
LCkXk
k=O
00
+ Co + L[(k
 (k  l)Ck]Xk
k=l
Thus
+ Co =
2C2
(k
+ 2)(k + 1)Ck+2 + (k
 l)kck+l  (k  l)Ck = O
and
C2
1
= "2co
kCk+1
(k  l)Ck
ck+2 = k+2  (k+2)(k+
1)'
Choosing
Co =
1 and
C1
Co
= 1,2,3, ....
= Owe find
C2
= O and
Cl
="2'
C4 = O
6'
C2 = C3 = C4 = ... = O. Thus,
C3
= 1 we obtain
Y = el ( 1 +"2x
1 3
+ 6x
+... ) + e2X
and
y' =
The initial conditions imply
y =
el =
el
(x
2 and Cz = 6, so
2ex
= O.
6. 1
Exercises 6. 1
26. Substituting y = L~=O cnxn into the differential equation we have
(x+1)y"
 (2  x)y'
+y
00
L n(n
00
n=2
00
L n(n
 l)cnxnl
 1)cnxn2
n=2
k=n2
00
00
L(k + l)kck+lxk
k= 1
n=l
k=nl
00
L ncnxnl
 2
L ncnxn
n=O
k=n
n=l
'..'
k=nl
'....'
k=n
00
L(k + 2)(k
k=O
00
k=O
L cnxn
00
00
k= 1
k=O
L kCkXk + L Ckxk
00
= 2C2 
+ Co +
2Cl
L[(k + 2)(k
k=l
Thus
2C2  2Cl
(k
+ Co = O
=O
and
C2 =
2'co
ci 
Ck+2 = k
+ 2 Ck+l
+ 2 Ck,
= 1,2,3, ....
Cl =
2"
C4
1
12
=
1 we obtain
C2 =
1,
C3
= O,
e4
1
=4
el
121314x
( 1  x
2
6
y'
el
(x
+ x
+...
12
+ e2 ( x + x 214 x
+ ... )
and
 ~x2
2
el
+ ~x3 + ... ) + e2
3
= 2 and
2 (1  ~x2
2
')
~x3
6
1
3
= 2  x  2x  x
e2 =
(1 + 2x
 x3
+ ... ) .
1, so
+ x5
12
+ ....
254
Exercses
27. Substituting y
+ 8y
y"  2xy'
00
=L
n(n  1)Cnxn2
 2L
n=2
00
nCnxn
+8 L
cnxn
n=O
~
k=n
n=l
''
k=n2
k=n
00
00
= L (k
+ 2)(k + 1)Ck+2Xk
 2L
k=D
00
+8L
kCkXk
Ckxk
k=O
k=l
00
= 2C2 + 8co
2k)Ck]Xk = O.
k=l
Thus
2C2 + 8co
(k
=O
+ 2)(k + 1)Ck+2 + (8
 2k)Ck = O
and
C2 =
4co
2k  8
Ck+2 = (k
Choosing
Co
= 1 and
Cl
+ 2)(k + 1) Ck,
k = 1,2,3, ....
= Owe find
C2 =4
Por
Co
= O and
Cl
C3
C4
4
=
C6
= ca =
C5
C7
= ... =
3
ClO
= ... = O.
= 1 we obtain
C5 =
10
( 1
and
Y , = el ( 8x
4 x 4)
4x 2 + '3
+ e2 ( x
16 x 3) + e2 ( 1 +3
255
1
 x 3 + 10x
1
3x +:2x
+ ... )
4 +... ) .
6. 1
Exercises 6. 1
The initial conditions imply Cl = 3 and C2 = O, so
y
28. Substituting
= 3 (14x2
+ ~x4) =
+4x4.
(x2
3 12x2
+ l)y" + 2xy' =
L n(n
00
 l)cnxn
n=2
+L
00
n(n  1)cnxn2
n=2
k=n
L k(k
2ncnxn
n=l
'v'
k=n2
00
+L
k=n
00
00
k=a
k=l
+ L (k + 2)(k + 1)Ck+2Xk + L
 l)ckxk
k=2
2kCkXk
00
Thus
k(k
=O
and
Ck+2 =
Choosing ca = 1 and
Cl
= Owe find
C3
 k + 2 Ck, k = 2, 3, 4, ....
= C4 = Cs
C3
1
=_
1
Cs =;:
C7
=_
7
and
Y
, = Cl (1 
x2 + x4  x6
256
+... ) .
Cl
= 1 we obtain
Exercises
The initial conditions
imply ca = O and
Cl
71x 7 + ....
29. Substituting
Y 11
= 1, so
31x 3 +"51x
Y =x 
+ (sism x ) y
= ~
L.. n ( n  1) Cnx n2
n=2
~ [2C2 + 6C3X
= 2C2 + (6C3
+ (x
1
 x
6
3 + x1
s _...
120
1 ) x3
"6ca
 ~ca) x3
+ ... = O.
Thus
6C3 + ca = O
+ ci
12c4
20cs
and
=O
1
Ca = O
6
+ C2 
C2 = O
C3
1
= Ca
C4
1
= 12Cl
1
Cs =  20C2
Choosing
ca
1 and
Cl
= O we
1
120ca.
find
C4
= O and
Cl
C2 =
and so
011.
6. 1
Yl
1 3
1  x
6
1
Cs = 120
= O,
1 we obtain
O,
C3
= O,
+ ...
1
C4
= 12'
Cs = O
are
1
+ x
120
and
257
Y2 =
X 
x + ....
12
+ ...]
Exercises 6. 1
30. Substituting y = I:;:;O=o
cnxn into the differential equation we have
00
y"
+ eXy'
 y =
I:n(n
 1)Cnxn2
n=2
(1 + x + ~x2
2
= [2C2 + 6C3X
+ ~x3 + ...)
(Cl
+ ci
 ce)
x2
x2
+ ... = O.
Thus
2C2 + Cl
Co
=O
6C3 + 2C2 = O
1
12c4 + 3C3 + C2 + 2Cl = O
and
C2 =
Co 
Cl
1
C2
3
1
C4 = C3
4
C3 =
Choosing
Co =
1 and
Cl
+ C2
12
1
24
Cl
= Owe find
1
C2 =
f Cnxn
n=O
Co = O and Cl =
2'
C4 = O
1 we obtain
1
C3
= 6'
1
24
C4 =
and
1
Y2 = x  x
2
258
+ 1x 3
6
1 4
  x
24
+ ....
Exercses
31. Substituting y
y"  xy =
00
L n(n
 1)Cnxn2 
n=2
L Cnxn+l
n=O
~
k=n+l
k=n2
00
L (k + 2)(k
00
+ 1)Ck+2Xk
k=O
L Ck_lXk
k=l
00
= 2C2 +
L [(k + 2)(k
+ 1)Ck+2  Ck_l]Xk =
l.
k=l
Thus
(k
+ 2)(k + 1)Ck+2 
Ckl
=O
and
1
C2 =
k = 1,2,3, ....
Ck+2= (k+2)(k+1)'
C2 =
2'
C3 = '6CO
1
C4 = 12cl
1
Cs = 20C2 = 40
and so on. The solution is
Y
1213141
Substituting y =
40 Cs +
14) +...
+ Cl ( X + 12x
...
ls + ....
+ '212x + 40x
y"  4xy'  4y =
L n(n
00
 1)Cnxn2 
n=2
L 4ncnxn
n=l
...._,..._.
k=n2
259
k=n
00
L 4Cnxn
n=O
k=n
'v'
6. 1
Exercses
6. 1
00
00
+ 2)(k + 1)Ck+2Xk
= (k
00
4kckXk
k=O
4CkXk
k=l
k=O
00
= 2C2  4co
+ [(k
+ 2)(k + 1)Ck+2
+ l)cklxk
 4(k
k=l
=e
= 1 + L..... k'x
k=l
k
o
Thus
2C2  4co = 1
(k
+ 2)(k + 1)Ck+2 
4(k
+ l)Ck =
1
k!
and
C2 =
"2 + 2co
1
Ck+2 = (k
+ 2)! + k + 2 ck,
= 1, 2, 3,
C2 =
"2 + 2co
1
3!
4
3
1 4
+ ct
3! 3
C3
=  + ct
C4
=
C5
C6
= 6! + "6C4 =
C7
=
1 4
1
1
13
+
C2 =  +  + 2co = 4! 4
4! 2
4!
1
51 + 5"C3
1
= 5! + 5 3!
o
16
+ 15c1
4 13 8
4! + "6co
51 + 6
1
+ 2co
4 17
o
64
17
16
261
51 + 15C
51 + 3co
409
64
260
Exercises
Y = Ca + CIX + ( 1 + 2ca ) x 2
2
(261
6!
+ ~CO)X6 +
3
4
= Ca [ 1 + 2x 2 + 2x 4 + x
(409
7!
6
+ x21 2 + x3!1
32. We identify P(x)
x4/5! "',
= Oand
for [z]
+ x13
4!
Q(x)
1 + CI
4)
+ ( "1
3. 3
X 3  (13
,+
4.
+ ~CI)Xi
+ (17
,+
5.
= sin x [.
16)
CI
15
X5
+ ".
105
2ca ) x 1
6. 1
15
+ x409
7!
105
+".
+" ..
differential equation.
33. The differential equation xy" = O has a singular point at x = O. It also has two solutions, YI = 1
and Y2
34.
If x
= z, that
are analytic at x
= O.
< O and the graph o a solution curve is concave down. Thus,
= xy
whatever portion o a solution curve lies in the first quadrant is concave clown. When x
y < O, y"
= xy
> O, so whatever portion o a solution curve lies in the fourth quadrant is concave
up.
35. (a) Substituting y = L~=a Cnxn into the differential equation we have
00
y"
> O and
+ xy' + y
L n(n
00
 1)cnxn2
n=2
n=l
k=n2
00
00
+ L ncnxn + L Cnxn
L (k + 2)(k
k=a
n=a
k=n
'v'
k=n
00
00
k=l
k=
Thus
2C2 + Ca = O
(k
and
1
C2 = Ca
2
261
=O
Exercises 6. 1
C2 =
C3
Co
= Oand
Cl
= Cs = C7 = ... = O
= 1 we obtain
C2=C4=C6==O
C3
= 
1
2
=3
3!
_~(\!2)
=_
\!2
6.;!. 2
= k=O
L......t
Yl
(_l)k
2k
2k . k' x
and
4
2
4
2
N=2
4
2 N=10
N=6
4 x
4
4
4
4
262
Exercises
(e)
yl
y2
4/
2'
6. 1
2
2
4
4
The graphs of Yl and Y2 obtained from a numerical solver are shown. We see that the partial
sum representations indicate the even and odd natures of the solution, but don't really give a
very accurate representation of the true solution. Increasing N to about 20 gives a much more
accurate representation on [4,4].
(d) From e" = 2:.'{'=oxk/k!
= Lk=O(1)kx2k/2kk!.
Frorn
= (
~ (_l)k
6
2kk!
2k) (~
x
6 (2k + 11)2kk!
2k+l)
x
+ (cos x)y
= 2C2 + 6C3X
+ (1  .:._ +.:._  .:._+ ... )(co + CJx + C2x2 + C3x3 + C4x4 + csxS + ... )
2!
(2C2
4!
6!
+ (30C6 + C4 + ~co
24
 ~C2)x4
2
263
Exercses 6. 1
Then
and
which gives
C6
= co/80 and
C7
= 19cr!5040.
Thus
Yl(X) = 1 
2X + 12X
1
 80X
Y2(X) = X 
6X + 30X
19
 5040X
+ ...
and
7
+ ....
(b) From part (a) the general solution of the differential equation is 'Y =
y(O) = Cl + C2' 0=
Y = Yl
(e)
and y/(O) =
Cl
+ Y2
= 1+ x
"2x
Cl .
2
CIYl
6x + 12x + 30x
1
 80x
19
 5040x
+"'.
4
2
x
2
4
(d)
Then
+ C2Y2.
6
4 2
2
2
4
4
264
Exercises 6.2
Exercises 6.2
= 3; regular
singular point: x
3
= 1;
= 0,3
we see that Xo
= 1 and
Xo
, +y=O
x
y " +y 5
x1
x+1
1 are regular singular points. For Xo
+ 5(x
 l)Y'
x(x  1)2
Y = O.
x+1
1). For Xo = 1 the differential equation can be
x1
In this case p(x) = (x + l)/(x  1) and q(x) = x(x
+ 1).
y' + 7xy = O
x
we see that Xo = O is a regular singular point. Multiplying by x2, the differential equation can be
y"
= O.
13.
r(r  1)
+ ~r
3
 ~ = r2 + ~r  ~ = ~(3r2
3
3
3 3
265
+ 2r  1) = ~(3r
3
 l)(r
~+ x
and q(x) =
+ 1) = O.
Exercises 6.2
The indicial roots are
k and
1. Since these do not differ by an integer we expect to find two series
= 10x.
+r
r(r  1)
= r2 = O.
The indicial roots are Oand O. Since these are equal, we expect the method of Frobenius to yield a
single series solution.
15. Substituting y
= L~=o cnxn+,.
2xy"  y'
+ 2y
+ L [2(k + r
 l)(k
+ r)ck
 (k
+ r)ck + 2Ck_l]Xk+,.1
k=l
which implies
2r2  3r
= r(2r  3)
=O
and
(k
The indicial roots are r = Oand r
+ r)(2k + 2r
 3)Ck + 2Ckl
= O.
Ck =
2Ckl
 k(2k _ 3)'
k = 1,2,3, ... ,
and
= 2co,
Cl
For r
= 3/2
C2
= 2co,
and
ci
= Sco,
C2 =
= 1,2,3,
... ,
2
35 Co,
 x4
945
16. Substituting y =
2xy"
35
+... ) .
terms, we obtain
00
+ L [2(k + r)(k + r
k=2
= O,
266
 l)Ck
O,
Exercises 6.2
which implies
2r2
+ 3r = r(2r + 3) = O,
(2r2 + 7r + 5)
= O,
Cl
and
(k
The indicial roots are r = 3/2
Ck2
Ck =  (2k  3)k '
k = 2,3,4, ... ,
and
1
C2
C3 = O,
= "2co,
Ck2
 k(2k + 3) , k
= 2,3,4, ... ,
and
1
C3 = O,
C2 = ~ 14co,
C4 =
1
616co.
Clx
1
1  x
2
= :;~=o Cnxn+r
+ ~yl + Y =
2
3/2
+ x1
40
+... ) + C2 ( 1 
x1
14
+ x1
616
+... ) .
k=l
=0,
which implies
4r2  ~r = r (4r 
+ r)(8k + 8r
+ Ckl
=O
and
1
"2(k
 7)q
= O.
The indicial roots are r = O and r = 7/8. Por r = Othe recurrence relation is
Ck =
2Ck_l
k(8k _ 7)'
1,2,3, ... ,
and
Cl
C3= Co
= 2co,
459
267
Ckl]
xk+rl
Exercises
6.2
ck=(8k+7)k'
1,2,3, ... ,
and
2
Cl
= 15co,
345co,
C2 =
C3
4
=  32,085 Co
2
+ x
9
 x4
459
+... ) + G2X7/8
1  x2
15
+ x2
345
 x 4
32,085
+... ) .
= :~=o Cnxn+r into the differential equation and collecting terms, we obtain
2x2y"  xy'
+ (x2 + 1) y
= (2r2  3r
+ 1) coxr + (2r2 + r)
ClXr+l
00
L [2(k + r)(k
+r 
+ r)ck + Ck + Ck_2]Xk+r
l)Ck  (k
k=2
= 0,
which implies
2r2  3r
+ 1=
(2r  l)(r  1) = 0,
(2r2 +r)
Cl =
+ 2r 
+ l]Ck + Ck2 = O.
0,
and
[(k + r)(2k
3)
Cl
k=2,3,4,
... ,
and
1
C2
= "6co,
C3
= 0,
1
C4
= 168 Co
k = 2,3,4, ... ,
and
C3
= 0,
1/2 (
1
1  x
6
+ x1
168
x
10
+ x1
360
+... ) .
Exercises 6.2
19. Substituting y
3xy"
= L:;:O;o Cr,xn+r into the differential equation and collecting terrns, we obtain
+ (2 
x)y'  y
(3r2
r) coxrl
00
+ 2:J3(k + r
 l)(k
 (k
+ r)ck_1Jxk+rl
k=l
= O,
which implies
3r2
r = r(3r 1) = O
and
(k
The indicial roots are r
+ r)(3k + 3r
 l)Ck  (k
+ r)ckl
= O.
k=1,2,3,
... ,
and
1
ci
For r
= 2'co,
Ckl
Ck=
3k'
2 3
'=1"
, ... ,
and
1
C]
= '3co,
C2 =
1
18co,
C3 =
1
162co.
1
Cl ( 1 + x
2
20. Substituting y =
+ x1
10
+ x1
80
1
1 + x
3
+ x1
18
+ x1
162
+".
L:;:O=o Cr,xn+r into the differential equation and collecting terms, we obtain
= O,
which implies
and
[(k
+ r)(k + r
 1)
269
~J
q
ckl
= O.
Exercses 6.2
The indicial roots are r = 2/3 and r = 1/3. For r = 2/3 the recurrence relation is
3Ckl
Ck = 3k2 + k '
k = 1,2,3, ... ,
and
3
Cl = 4co,
C3
= 560co.
k = 1,2,3, ... ,
and
3
Cl = 2'co,
C3
= 160co.
2/3 ( 1
+ x3 + x9
56
+ x9
560
20
+ x9
160
+... ) .
21. Substituting y = :L~""ocnxn+r into the differential equation and collecting terms, we obtain
00
2xy"  (3
+ 2x)y' + y
+ "L[2(k + r)(k + r
 l)Ck
k=l
 2(k
+r
5r = r(2r  5)
=O
 3(k
+ r)ck
 l)Ckl
+ ck_dxk+rl
=0,
which implies
2r2
and
(k
The indicial roots are r
+ r)(2k + 2r
 5)Ck  (2k
+ 2r
 3)Ckl
(2k  3)Ckl
Ck = k(2k _ 5) ,
k = 1,2,3, ... ,
and
1
Cl = 3co,
For r = 5/2 the recurrence relation is
Ck =
= O.
2(k + l)Ckl
k(2k + 5) ,
and
270
k = 1,2,3,. " ,
Exercises
6.2
= Gl ( 1 + x1
3
22. Substituting y
1 2  x
1 3
 x
6
4
1 + x
7
+ x4
21
+ x32
693
+... ) .
= ~=a cnxn+r into the differentia! equation and collecting terms, we obtain
f:
[(k
+ r)(k + r
 l)Ck
+ (k + r)ck
iq +
Ck2]
xk+r
k=2
=0,
which implies
r2  ~ = (r
(r2
D D
= O,
(r 
+ 2r + ~) ci
= O,
and
[(k
= 2/3
and r
Ck =
+ r)2
 ~] q + Ck2
= 2/3,
so
= O.
= O. For r = 2/3
9Ck2
3k(3k  4) ,
k = 2,3,4, ... ,
and
C3
= O,
C4 =
Ca
128
3k(3k
+ 4) ,
k = 2,3,4, ... ,
and
C3
= O,
 Ca
1280
.
,
C4 
3 2 + x9
Glx 2/3 ( 1  x
4
128
1  x3
20
+ x9
1,280
+... ) .
Exercises
6.2
23. Substituting y = L~=O cnxn+r into the differential equation and collecting terms, we obtain
9x2y"
+ 9x2y' + 2y
= (9r2  9r
+ 2) Coxr
00
+ ~[9(k
k=l
= O,
which implies
9r 2
9r
+2 =
(3r  1)( 3r  2)
=O
and
[9(k
and
ci
For r
= O.
(3k  2)Ckl
k(3k  1) ,
Ck = 
l)Ckl
= 2"co,
= 1,2,3, ... ,
C3
7
120
= Co.
Ck=and
k = 1,2,3, ... ,
ci =
2"co,
C3
1
21
= Co
2x2y"
1/3 (
1
1  x
2
1 2  x7
+ x
5
120
+...
28
 x1
21
+...
= L~=O cnxn+r into the differential equation and collecting terms, we obtain
= (2r2
+r
1) coxr
00
= O,
which implies
2r2
+r
 1 = (2r  l)(r
+ 1)
=O
and
[(k
= O.
Exercises 6.2
The indicial roots are r
= 1
= 1/2. For r = 1
and r
Ck =
2Ckl
k(2k _ 3)'
= 1,2,3, ...
and
ci
= 2co,
C2
= 2co,
Ck =
and
= 1,2,3, ... ,
ci
= "5co,
C3 = Co
945
= Clx 1 (
25. Substituting y
xy"
4
1 + 2x  2x 2 + x
9
2
x
5
+ x2
35
 x4
945
+ ....)
= ~=o Cnxn+r into the differential equation and collecting terms, we obtain
+ 2y'
 xy = (r2
+ r)
+ (r2 + 3r + 2) Cl xr
coxrl
00
+ :L [(k + r)(k + r 
l)Ck
+ 2(k + r)ck
 ck_21xk+rl
k=2
= 0,
which implies
r2
+ r = r(r + 1) = O,
(r2
+ 3r + 2)
Cl
= O,
and
(k
The indicial roots are
rl
= O and r2
+ r)(k + r + l)Ck
= 1,
so
= O. For
Cl
Ck2
Ck = k(k
 Ck2 = O.
+ 1) ,
rl
k=2,3,4,
and
1
C2 =
3fco
C3 = es = C7 =
C2n = (2n
... = O
+ l)!co.
273
= Othe
... ,
recurrence relation is
Exercises
For
1"2
6.2
k = 2,3,4, ... ,
and
C3
= Cs = C7 = ...
C4
4TCQ
1
C2n = (2n)! Co .
The general solution on (0,00) is
Y=
e ~
2n
l;:O(2n+1)!x
[e1;:0~
(2n
= ;;;
1
x
+ l)!x
1
;:O(2n)!x
1 ~
2X
2n+ 1
ef
+ 2 n=O
2n
1 2n]
(2n)!x
+ xy' + ( x2
 ~) y = (1"2  ~) cQxr
f [(k + 1")(k +
1" 
l)Ck
+ (k + 1")ck 
~2
~Ck
+ Ck2]
xk+r
=0,
which implies
(1"2 + 21"
and
C1
= O,
[(k+T)4 .) 1] Ck+Ck2=0.
1"1
= 1/2 and
1"2
= 1/2, so
Cl
Ck2
Ck =  k (k + 1) ,
274
= O. For
1"1 =
= 2,3,4, ... ,
Exercses
and
1
C2
= CO
3!
C3
Cs
C7
= ... = O
(_1)n
C2n = (2n + 1)! en.
For r2
k = 2,3,4, ... ,
q=k(k1)'
and
C2 =
C3
1
cO
21
= Cs
C7
= ...
=O
1
C4 =
4!cO
(_1)n
C2n = (2n.)1 Ca
The general solution on (0,00) is
y
CIxl/2
n=O
= CIXI/2
n=O
f (2n(_l)n+ 1)!x2n+1 + C2
XI/2
n=O
f (_l)n
(2n)!
x2n
n=O
I::~=o Cnxn+r into the differential equation and collecting terms, we obtain
00
xy"  xy'
+ y = (r2  r) coxrl +
.L [(k + r + l)(k
k=O
which implies
r2  r = r(r  1) = O
and
rl
= 1 and r2 = O. For
rl
kci.
Ck+1 = (k+2)(k+l)'
k=0,1,2, ... ,
275
6.2
Exercises 6.2
and one solution is YI = CoX. A second solution is
Y2 = X
J J
dx
e x2
dx = x
J :2
x
= xJ
1 (
dx = x J x2 1 + x
+ !"x2 + ...) dx
1 ')
+ x1
2
12
= x ln x  1 + x
=x
4!
+ x1
72
+.!.x
2
dx
72
+ ....
28. Substituting y
y"
= I:~="o cnxn+r into the differential equation and collecting terms, we obtain
3
+ ;yl
 2y = (r2
+ 2r)
coxr2
+ (r2 + 4r + 3)
CIXr1
00
L [(k + r)(k
+r
 l)Ck
+ 3(k + r)ck
 2ck_21xk+r2
k=,,2
= O,
which implies
+ 2r
r2
(r2
(k
The indicial roots are TI = O and
+ 2) = O
= r(r
+ 4r + 3)
CI
+ r)(k + r + 2)Ck 
r2 =
2Ck2 = O.
2, so c = O. For
2Ck2
Ck = k(k + 2)'
rl = O the
recurrence relation is
k = 2,3,4, ... ,
and
1
C2
= ::tco
C3= C5
= C7= ... = O
1
C4
= 48co
1
1,152
C6= Co
The result is
Yl
1214+
1 x
= Co( 1 + x
4
48
276
+ C6
+
1,152
oo'
Exercises 6.2
A second solution is
Y2 = YI J
= YI
= YI J
e J(3jx)dx
2
dx = YI
YI
dx
x3 (1 + Ix2
+
~x4
2
48
+ _!_x6
+ ...) = YI
576
~X3
576
+y
= ~YIlnx
[__
dx
x3
(1 + Ix2
+ _'!_x4
+ ...)
4
48
+ ...) = YI
+ _2'_x2 _
1_
2x2
[ __
1 (
1 2
x3 1  2x
1_  ~
2x2
lnx
7 4
19 6
)
+ 48
x + 576x + ...
+ _2'_x2 96
~x4
2,304
+ ...]
+ ...] .
~x4
2,304
96
Cr,xn+r
xY"
+ (1
x)Y'  y
= T2cOXr1 + L[(k
+ T)(k + T 
l)Ck
+ (k + T)Ck 
(k
+ T)ck_Ilxk+r1
k=O
which implies
T2 = O
and
TI =
CkI
k
'
k = 1, 2, 3, ....
One solution is
YI = Co ( 1 + x
+ x21213+ IX
+... )
3.
= coe
A second solution is
Y2 = YI
e J(ljxl)dx
e2x
= e xJ1( 
1 x
dx = eX
+ x1213 2
= eX[l nx  x + x1
J eXIx
J 1
e2x dx = e" ;;exdx
x
3!
2  x1
2 2
3 .31
+ ... )
dx
3!
3 + ... ] = e xl nx  e x~(_l)n+1
L...,
x n.
n=1 n n!
277
= O,
Exercises 6.2
30. Substituting y
xY"
+ y' + y
+ [(k + r)(k + r
= r2cQxr1
 1)Ck + (k
+ r)Ck + Ck_l]Xk+rl = O
k=l
= O and
which implies r2
"i
One solution is
Yl
= ca
( 1 x
+ 22x
1
 (3!)2x
A second solution is
Y2
e f(l/x)dx
= u.
= Yl
J x ( 1  2x + 2x2 
dx
Yl
J
u: J (~ +
1(
5
;; 1+2x+2'x
= Yl lnx
= y1ln x
+ (4!)2x
2 + ~x
ca
i: (1)n
(n!)2
n
x .
dx
(1  x + ix2
 3~x3
+ ...)
35
+ 288x4
5
gX3
23
+gX
 ...
4
)
+ 677
288x + ...
dx
288
27
+ x677
<1
1,152
5 2 23
+ x
+ x
( 2x
...
dx
5 2 23
+ 2x + x
+ x
+ Yl
J
X
=Yl
=
2
Yl
27
+ x677
+ ... ]
4
1,152
+... ) .
31. Substituting y
= I:~=a
xy"
+ (x
 6(k
+ r)ck
 3Ck_dxk+r1
which implies
r2
7r
= r(r
278
 7)
=O
O,
1)Ck + (k
+r 
1)Ckl
Exercises 6.2
and
(k + r)(k + r  7)Ck + (k + r  4)Ck1 = 00
The indicial roots are r1
+ (k + 3)Ck1
=0,
k=l,2,3,ooo,
or
k
Taking co
= 1,2,3,
=f. O we obtain
1
C1 = Co
C2
5
18cO
1
C3 = 6"co,
and so on. Thus, the indicial root r1 = 7 yields a single solution.
relation is
k(k  7)Ck
+ (k
 4)Ck1 = O,
k = 1,2,3,
Then
6C1  3co = O
12c3C2=0
12c4 + OC3 = O
===}
C4
=O
+ C4 = O
===}
Cs
=O
=O
===}
C6
=O
OC7 + 3C6 = O
===}
C7 is arbitrary
lOcs
6C6 + 2cs
and
k4
Ck =  k(k _ 7) Ck1,
279
k = 8,9, lO,
Exercises 6.2
Taking Co 1= O and C7 = Owe obtain
1
Cl = Co
2
1
C2= lOcO
1
C3 = Co
120
C4
Taking
= es = C6 = ... = O.
Co = O and C7 1= O we obtain
ca = C7
2
Cg =
ClO
1
=  36c7'
= ~=o
x(x  l)y"
+ 3y' 
cnxn+r
5
36c7
Y2 = X
and
1 a
x
2
+ x5
36
1
 x
36
10
+ ....
2y
00
+ [(k + r
 l)(k
+r
 12)Ck_l  (k
+ r)(k + r
k=l
= O,
which implies
4r  r2 = r( 4  r)
and
(k
The indicial roots are
+ r)(k + r
rl
= 4 and
(k
 4)Ck + [(k
r2
+r
 l)(k
= O. For rl = 4 the
+T
or
k=1,2,3,.
280
2)  2]Ckl
= O.
recurrence relation is
2]Ckl = O
 l)Ck
Exercises 6.2
Taking
Co =1=
Owe obtain
CI =
2co
C2
= 3!co
C3
= 4!co,
TI
T2
is
k(k
 4)Ck
+ k(k
 3)Ck1
= O, k = 1,2,3, ... ,
or
(k4)Ck+(k3)Ck_1
=0,
k=
1,2,3, ....
Then
3CI 
2co = O
2C2  CI = O
C3
+ OC2
OC4
=O
+ C3 = O
=:>
C3 =
=:>
C4
is arbitrary
and
Ck =
Taking
Co =1=
O and
C4 =
(k3)Ck_1
c4
k = 5,6, 7, ....
Owe obtain
2
C2
= Co
3
1
= "3co
C3 =
Taking
Co =
O and
C4 =1=
C4 = C5 = ... = O.
Owe obtain
CI = C2 = C3 =
= _1/x2
5364
Y2 = x 42+
x7
= _t2
Then
281
+ ....
Exercises 6.2
dy
dx
dy di
dt dx
_=
_
dy
dt
t 
and
Now
x4 ~d2
dx
+ Ay
1 (d2
t4 __J!_2
dt
+ 2t3 ddY ) + Ay
= 4"
d2 + 2 _}j_
d + Ay = O
i
dt
t dt
becomes
d2y
t d
(b) Substituting y
d2y
t d 2
dy
+ 2 dt + >..ty
= O.
+ (r2 + 3r + 2)Cltr
00
+ L [(k + r)(k + r 
l)Ck
k=2
= O,
which implies
r2
+ r = r(r + 1)
(r2
+ 3r + 2)
= O,
CI = O,
and
(k
r2 =
1, so
ACk2
k( k + 1) ,
CI = O.
For
C3
O.
rl = Othe
k = 2,3,4, ... ,
and
C2 =
A
3!
ca
= C5 = C7 = ...
282
=O
recurrence relation is
Exercises
For
r2 =
6.2
= 2,3,4, ... ,
and
.A
C2
= ca
2!
C3
= Cs =
C7
= ... = O
~[CI
t
n=
(V>:t)2n]
n=
y(X)CIX
34.
+ C2X
v>:
cos .
x
CI sin 
+ C2 cos v>:
a
v>:
= O
v>:
Slll
.~
sm
v>:
cosa
~
cosb
. v>:
v>:
v>: . ~
= sm  cos  cos  Sllla
b
a
b
=sin(~
 ~)
283
=sinv>:C~ba)
=0.
Exercises 6.2
This will be the case if
v0.
or
n ,'\ab
=b_
nAab
= L'
n = 1, 2, ... ,
or, if
The criticalloads
we have
= Clx[sin
v0. _
x
= C3X [Sin
and
Using C2
(1 1) = C4xsinn7fab ( 1
Yn(x) = C3xsinn7fab   L
(b) When n
= 1,
= 11, and
1, we have,
a) .
x
for C4 = 1,
Yl(x) = x sin 1.l7r
(1  ~) .
36. Substituting y = 2:~=o cnxn+r into the first differential equation and collecting terms, we obtain
00
x3y" + y = coxr +
L [Ck + (k +
7" 
l)(k + 7"
2)Ck_lJxk+r
k=l
It follows that Co
= Oand
Ck =
(k + 7"
l)(k + 7"
284
2)Ck_l'
= O.
Exercses
Substituting y
6.3
= L~=O Cnxn+r into the second differential equation and collecting terms, we obtain
00
x2y"
+ (3x
 l)y'
+y
+ 2: [(k + r + 1)2Ck 
= rco
(k
+ r + l)Ck+dxk+r
= 0,
k=O
which implies
+ r + 1)2ck
(k
If en
rco
 (k
+ r + l)Ck+l = O.
we obtain
Ck+l = (k+ l)Ck,
Letting
Co
is then y
= 1 we get
Cl
= 2,
= L~=O(n + l)!xn,
C2
= 3!,
C3
k = 0,1,2,
....
= O.
+ (b/a)xy' + (c/a)y
= O and identify ao
= b/a
+ a r + a =
or
ar2
+ bxy' + cy
+ (b 
a)r
= O.
Exercises 6.3
1. Since v2
+ C2Ll/3(X).
3. Since v2
+ c2 Yl (x).
= qJS/2(x)
+ c2Ls/2(x).
4. Since v2
= qJl/4(X)' + c2L1/4(X).
5. Since v2
6. Since v2
7. Since v2
8. Since v2
+ C2Yo (x).
= 4 the general solution is y = clJ2(X) + C2Y2(X).
= 2 the general solution is y = qh(3x)
+ C2Y2(3x).
= 1/4 the general solution is y = qJl/2(6x)
+ C2J_l/2(6x).
9. If y = xl/2v(x)
+ c = 0,
then
'21x3/2v(x),
y' = xl/2v'(x)
y" = Xl/2v"(x)
_ x3/2v'(x)
285
+ ~xS/2v(x),
4
Exercises
6.3
and
x2y"
+ 2xy' + x 2x2y
= x3/2v"
+ X 1/2v' + ( x 2x3/2
 ~X
1/2) v.
= xnJn(x)
+ ()..2x2
+ c2J_1/2()..X).
= c1J1/2()..X)
~)
V = O,
Then y = cx1/2J1/2()..X)
+ C2X1/2J_l/2()..X).
we find
and
+ 2nxn
+ n(n
J~
= xnl
 1)xn1 Jn
.O
+ (1 
2n)(xn J~
+ nxn1
Jn)
+ xn+l
Jn
= O.
nJ' nnx
nlJ
+ (1 + 2n)y' + xy
= xn1
[x2J~
= xn1
+ xJ~ + (x2
= O.
= )..v'xJ~()"x)
 n2) Jn]
+ ~xl/2Jv()..x)
and
286
Exercises 6.3
Substituting into the differential equation, we have
x2y" + (A2X2  v2 +
JXo
= o.
Therefore,
JX
= x1/2J1/2(X).
= xJ1(x).
= 1/2
we find
y = x1/2J_1/2(X).
14. From Problem 10 with n = 1 we find y
y = xJl(X)
X1j_1(X).
JX Jo(2x).
we find y = JX J3/2(X)
and y =
JX J3/2(X).
1 we find
xJl(X).
y = X1Jl(X)
= 3 we find
= X3j_3(X) = x3]s(x).
fa
00
vJv(x)
+ xJvl(X)
=
=_
(l)nv
(x)2n+v
+ v + n) '2
(::.)2n+v +
00
(l)nv
fan!r(l+v+n)
+ v) (::.)2n+v _
(1)n(2n
 n=O
L n.Ir( 1 + v + n )
00
00
(_l)n
(x)2n+vl
+ x ~o nlr(v + n) '2
nlf(l
00
(_l)n(v
fan!f(l+v+n)
+ n)
.2 (::') (::.)2n+vl
2
2
 xJv(x).
(b) The formula in part (a) is a linear firstorder differential equation in Jv(x).
An integrating
ddx[xvJv(x)] = XVJvl(X).
20. Subtracting the formula in part (a) of Problem 19 from the formula in Example 4 we obtain
0= 2vJv(x)
 xJv+1(x)  xJv1(x)
or
xJo(x)
dx[xJl(X)]
so
287
Exercises 6.3
22. From (14) we obtain Jb(x) = JI (x), and from (15) we obtain J(x) = Ll(X).
LI(X)
= JI
(x).
23. Since
1
r ( 1  "2 + n
(2n  1)!
(n _ 1)!22nI
we obtain
1/
= 1/2, we obtain
hj2(X)=Y;;; [2 (SinX
;cosx
with
1/
= 1/2
we obtain J_1j2(X)
1/
= xJ1j2(X) + xJ_3j2(X)
J_3j2(X) =
and with
ff
so that
so that
(CO;X + sinx);
(h)
0.5
1
1
y
'1
o.5
0.5
1
'1
L\ v ~
C>.3/2C">
5 'Cl10
""B
0.5 /""'\
20 x
v C>.
~ 5/2
C">
25. Letting
we have
dx = dx ds = dx [~ {k (_~) eo.tj2] = dx (_
dt
ds dt
dt a. Y ;;;,
2
ds
288
(k eo.tj2)
Y ;;;,
Thus Jb(x)
Exercises
and
d2x = .!!:._ (dX) = dx (::
dt2
dt dt
ds 2
(k eat/2) + .!!:._
Y :;
(dX)
ds
dt
(_
(k eat/2)
Y :;
Y :;
Y :;
ds2 dt
Y :;
ds2
Then
m d2x
dt2
+ keatx
= keat d2x
ds2
+ ma
2
If

eat /2 dx
dt
+ kea
t x = O.
Y :;
di
=O
a2 m
dx
+ s ds + s
x = O.
+ ~x1/2w
(~ax3/2)
and
+ aw'
+ ~aw'
Gax3/2)
(~ax3/2)
 ~x3/2w Gax3/2)
+ a2xy
= a [~x3/2wll
+ ~w' + (ax3/2
 4a~3/2) w] = O.
+ tw'(t) + (t2
 ~) w(t)]
= O,
t O.
=x
1/2w
> O is
6.3
Exercises 6.3
28. Setting y = ft JI (2ft)
y' = ft J~(2ft)
2e
2yx
le JI (2ft)
2yx
= J{(2y'X)
le JI (2y'X)
2yx
and
"
"(
"(
1,
ftJ1
el,
2yx)+2xJ1(2yx)
1
(e
 4x3/2 JI 2y X )
4x3/2Jl(2yx).
Jx
t2 [t2"
JI
[XJ~'(2y'X) +
le h(2y'X)
4yx
V; J;(2y'X)
+ (x 
+ y'X Jl(2y'X)
JI (2y'X)]
1) (]
2..
[ t2 J;' (t) + tJU t) + (t2 2t
1) JI
(t )l.
Since h(t) is a solution of t2y" + ty' + (t2  l)y = O, we see that the last express ion aboye is Oand
y = ft JI (2ft) is a solution of xy" + y = O.
29. (a) Using the expressions for the two linearly independent power series solutions, Yl(X) and Y2(X),
given in the text we obtain
satisfies
(1 x2)
290
(1 x2)
k = 2, 3, 4, ....
Exercises
6.3
15X)
8
d2y
. 2 d2y
dy
de2 = sm e dx2  cos dx '
and
d2y
sine de2
+ cose dy
de + n(n + l)(sine)y
[(
= sine
2 ) d2y
dy
1 cos e dx2  2cose dx
+ n(n + l)y ] = O.
That is,
d2y
( 1  x 2) dx2
32. Frorn
y = u/vx
dy
2xdx
+ n(n + l)y
= O.
we find
2xu'  u
y' = 2x3;C/""'2
and
Then
2 "
x y
+ xy +
(2
2)
4x5/2u"  4x3/2u'
x  u y=
4x
+ 3xl/2u
2xu'  u
2x1/2
(x2;2)u
xl/2
1 + x 2u + u
4
2u =
or
d2u
dx2
+ (1
;2
1/4)
x2
u= O.
+u
and cos x. Thus, we expect the solutions to oscillate with increasing frequency for larger x.
33. Rolle's theorem states that for a differentiable function f(x),
exists a number e between a and b such that f'(c)
JO(x) = h(x).
for which h(x)
f(b) = O, there
;
= O we have
Thus, if a and b are successive zeros of Jo(x), then there exists a e between a and b
= J6(x) = o.
291
Exercises 6.3
34. Since
Kh
= 4, k = 1, and
o:
= 0.1.
r 1dt=
K)O
l.
Then
+ c2Yo(10) =
ql(lO)
+ c2Y(10) =
1.
10
5
5
50
100
150
200
= O is
+ C2xl/2]_1/3 (~x3/2) .
= 1, x'(O.l) = ~ we find
1
292
Exercises 6.3
37. (a) Letting t
d2
dt2
+ ohe = O,
e/(o) = O,
e(L) = O,
where 0:2
= cltl/2Jl/3
+ C2tl/2J_l/3
Go:t3/2)
Go:t3/2)
= CJel(t) + C2e2(t).
= o implies
= O, and
el
e(t) = C2VtJl/3
=1=
so
Go:t3/2).
C2 =
O, in whieh
case
GO:L3/2)
= O,
e(t) = O, or J_l/3(~o:L3/2)
to bend when L is the length eorresponding to the smallest positive zero of Jl/3.
Mathematica,
Using
= 1.86635
implies
L =
(
3(1.86635)
)
20:
2/3
40g
= [9(2.6
X 107)7l'(0.05)4/4(1.86635)2] 1/3
4(0.28)7l'(0.05)2
+ (PL/M)y
76.9 in.
= O, we identify A
PL/M.
h(O)
+ C2.xYl
(2VPLX/M)
(2VPLX/M).
y = Cl.xJl
(b) From y(L) = Owe have y = JI (2L..fPM)
(2VPLX/M).
= O. The first positive zero of JI is 3.8317 so, solving
2 J3.6705X
L
) = Cl.xJl
293
(3.8317)
L.x
Exercises 6.3
(e) For
el
is shown.
0.2
0.1
x
0.2
39.
0.4
0.6
0.8
Jo
.Jo
JI
.J1
J2
.J2
2.4048
5.5201
8.6537
11.7915
14.9309
3.1153
3.1336
3.1378
3.1394
0.0000
3.8317
7.0156
10.1735
13.3237
3.8317
3.1839
3.1579
3.1502
0.0000
5.1356
8.4172
11.6198
14.7960
5.1356
3.2816
3.2026
3.1762
Successive zeros of Jn for n = O, 1, 2 appear to be approximately equally spaced for larger values
of x. Furthermore, this spacing appears to be approximately
can be used to determine that successive roots of Jo near x
it .
with difference 3.141583. This is consistent with the observation in Problem 32 that the frequency
of the solutions for large x is approximately 1/27r.
40. (a) Using a CAS we find
1 d1
P1(x) =
"2 dx1
P2(x)
1d22
212
222! dx2 (x  1) = "2(3x  1)
P3(X)
1 d3
2
3
233! dx3 (x  1)
= "2(5x  3x)
P4(x)
1 d4
244! dx4 (x2  1)4
= S(35x4
(x  1) = X
d5
30x2
+ 3)
616
d7
Exercises 6.3
(b)
P1
1
P2
1
0.5
1
0.5
1 x
1
0.5
1
P5
1
1
1
P7
1
P6
1
0.5
0.5
0.5
1
1
1
H(x) : O
P2(X) : 0.57735
P3(X) : O, 0.77460
P4(X) : 0.33998, 0.86115
P5(X) : O, 0.53847, 0.90618
P6(X) : 0.23862, 0.66121, 0.93247
P7(X) : O, 0.40585, 0.74153 ,0.94911
PlO(X):
The zeros of any Legendre polynomial are in the interval (1, 1) and are symmetric with respect
to O.
295
1. The interval of convergence is centered at 4. Since the series converges at 2, it converges at least
on the interval [2,10).
at 7 it does not converge, at O and 7 it does converge, and at 10 and 11 it might converge.
2. We have
x3
sin x
x
f ()x ==
cosx
3. Solving x2  2x
+ 10 =O we obtain
radius of convergence is
4. Setting 1  sin x
1
x5
6 + 120 x
'2 + 24 
x3
=x+++
2x5
...
J11, which
x =1
11  J111 = J11
...
15
.. .
1.
= O we see the singular points closest to O are 311'/2 and 11'/2. Thus, the minimum
+ y' + y
x2)y"
+ 3)y" + y = O has
= 3.
2xy"
+ y' + y =
(2r2  r) coxrl
+ L [2(k + r)(k + r 
l)Ck + (k
+ r)ck + ck_llxk+rl = O
k=l
which implies
2r2  r = r(2r  1) = O
and
(k
+ r)(2k + 2r 
l)Ck + ckl
so
Cl
= O.
= Co,
C3
= Co
90
+ 1) ,
= 1,2,3, ... ,
so
C3
296
1
630
= Co
( lx+6x
Y1=Cl
90X1
+ ... )
and
Y2 = C2X
1/2 (
'31x + 301 x 2 
1
1 3
630x +...
y"  xy'  y
00
00
00
+ 2)(k + 1)Ck+2Xk 
= L(k
k=O
00
L kCkXk  L Ckxk
k=l
k=O
00
= 2C2  Co
Thus
2C2  Co = O
(k
+ 2)(k + 1)Ck+2 
(k
+ l)Ck = O
and
1
C2
ck+2
Choosing ca = 1 and
C1
= '2co
= k + 2 Ck,
k = 1, 2, 3, ....
= Owe find
1
2
C2
=
C3
= C5
C4
=
C6
=
= C7
= ... = O
1
8
1
48
297
(k
+ l)CkJXk = O.
ca = O and Cl =
1 we obtain
C2 = C4 = C6 = ... = O
1
3
1
=15
C3 = 
C5
1
= 105
C7
= 1 + 2x + S"X + 48 x + ...
and
Y2=X+3X
1
+15x
1
+105x
+ ....
9. Substituting y
00
+ 3y =
(x  l)y"
(2C2
+ 3ca) + L (k
 l)(k 
2)Ckl
k(k  l)Ck
+ 3Ck_2]xk2 = O
k=3
which implies
C2 = 3ca/2
and
ce=
Choosing
ca =
1 and
Cl
(k  l)(k  2)Ckl
k(k1)
k = 3,4,5, ....
= O we find
3
C2 =
+ 3Ck2
ca = O and c
2'
C3 =
2'
C3 =
2'
5
8
C4 =
= 1 we obtain
C2
= O,
C4 = 
301354
= Cl ( 1 + x~ + x
2
2
+ x + ....)
8
and
Y2 = C2 (x
10. Substituting y
y"  x2y'
which implies
C2 = O, C3 = ca/5,
and
k1
k = 1,2,3,
298
....
 (k  l)cklxk+l
=O
Co
= 1 and
co
Cl
= Owe find
= O and
Cl
C3
1
=
C4
= C7 = ClO = ... = O
Cs
= Cs = Cll
C6
1
=_
= ...
=O
90
= 1 we obtain
C3
= C6 = Cg = ... = O
C4
C7
Cs
Cs
= Cll = ...
ClO
= ... = O
=O
= :;~=oCnxn+r
(1  ~x3 
:ox6  ... )
and
Y2 = ClX.
3r)coxrl
.L [(k + r)(k
+ r  3)Ck
k=1
 (k
+r 
3)Ck_Ilxk+rl
= O,
which implies
r2
3r = r(r  3) = O
and
3)Ck1 =
rl
O.
k(k  3)Ck  (k 
3)Ckl
= O,
= 1, 2, 3, ....
Then
Cl 
OC3 
Co
OC2
=O
=O
==>
C3
is arbitrary
and
k = 4,5,6,
299
....
C2 = 2"CO
C3 = C4 = Cs
= ... = O.
1
C4 = C3
6
4!C3
Cs
1
6
5 . 4 C3 = 5! C3
1
C6 = ~C3
= 6!C3,
1 2
= 1 + x + 2"x
and
Y2 = x 3 + x6 4 + x6
4!
5!
+y
= O.
Thus
Co + 2C2 = O
300
c? = Co
C3
= "6Cl
C4
=O
Cs =
C6 =
Choosing
Co
= 1 and
Cl
Co
1
1
C4  C2
6
360
= Owe find
= O and
"2'
Cl
= 1 we find
C2
1
10 C3
C2=0,
C3
= O,
C3="6'
C4
= O,
= O,
Cs
C4=0,
Cs
=  60'
es
=O
1 2
1
= 1  x
+ x
+...
= x  1x 3  x1
2
720
6
L~=O Cnxn into the differential equation we have
00
y"
Y2
and
+ xy' + 2y
2:= n(n
00
 1)Cnxn2
+ 2:=
n=2
00
2:= (k + 2)(k
k=O
60
+ ....
00
ncnxn
+ 2 2:= Cnxn
n=l
k=n2
n=O
k=n
'v"
k=n
00
00
k=l
k=O
Ckxk
00
= 2C2 + 2co
Thus
2C2 + 2co = O
(k
and
C2
= Co
ck+2 =  k
+ 1 Ck,
301
k = 1,2,3, ....
= O.
Cl
= O we find
C2
= 1
C3
C4
1
=
C6
=
Cs
C7
= ... =
C6
= ... =
15
C4
C3 =
Cs = 
8
1
C7=

48
ca ( 1  x 21416)
+ 3"x  15x
+... + cl
(131S
X "2x
+ B"x
17
 48x
+ ...)
and
Y,
= ca (
2x
+ x4332S 
= 2 we find
y = 3  2x  3x
14. Substituting y
+... ) + Cl
x
Co
3254
1  x
2
= 3 and
Cl
16
+ x + x 41S:x
23
+ x
8
= 2.
 5"x
+ ...)
 x76
48
+ 2417 x + ....
(x
+ 2)y" + 3y =
L n(n 
CXJ
+2
l)cnxnl
n=2
L n(n
+3
n=2
k=nl
L (k + l)kck+lXk
k=l
L cnxn
n=O
k=n
'v'
k=n2
00
00
 1)cnxn2
CXJ
CXJ
+ 2 L (k + 2)(k + 1)Ck+2Xk + 3 L
Ckxk
k=O
k=O
CXJ
Thus
4C2
(k
+ l)kq+l
+ 3ea
=O
=O
= O.
3
4
Co
k
Ck+2 = 
Choosing
Co
1 and
Cl
2(k
+ 2) Ck+l
2(k
k = 1,2,3, ....
+ 2)(k + 1) cu,
= Owe find
3
C2
=4
C3
=
1
8
1
C4 =
16
9
320
Co
= O and
Cl
=1
we obtain
C3
=4
C4 =
16
es = O
and so on. Thus, the general solution is
co
3213
( 1 :x + S"X
+ 1614x 
9
320x
s)+...
and
y,
Setting y(O)
21394)
= ce (33
2"x
+ SX +:x  64x +...
Co
= O and
+ Cl
(31213
:x
+:x +... ) .
Cl
problem is
Y
1 3
= x  x
+ x1
16
+ ....
11
(1 cos x ) y , + xy
x
= O,
is analytic at x
= O, we conclude
that x
= Ois an
303
y" +(eX  x1  x )y
=O
=++_
?(
X)
x: eX _ 1 _ x
2x2
x4
3
+ x3
18 + 270 
= 2x 
... ,
I.:~=Ocnxn
00
y"
+ x2y' + 2xy
2: n(n
00
n=2
2: (k + 2)(k
k=O
Cnxn+l
n=O
k=n+ 1
n=l
k=n00
00
+ 2: ncnxn+l + 22:
 1)cnxn2
~~
2
k=n+ 1
00
+ 1)Ck+2xk + 2: (k
00
+ 22: Ck_lXk
 l)Ck_lXk
k=2
k=l
00
20C5
(k
and
+ 4C2 = 10
= 4,5,6, ... ,
5
C2 = 2
1
C3=CO
C4
1
Cl
C5 = ~ 
~C2 = ~  ~
1
ck+2 =  k
+ 2 Ckl
304
G)
=O
+ 1) = 32.
1
2! co + 32.2'
1
1
C7 = C4 = Cl
7
47
Cs = Cll = C14 = ... = O
1
1
1
C9 = gC6 =  33. 3! co  33.3'
ClO =
1
10C7 =
1
4 . 7 . 10 Cl
C12 =  12 C9 = 34 . 4! Co + 34 . 4!
Cl3 =
1
1
13 Co = 4. 7 . 10 . 13 Cl
y=co
+ci [x'4x
1 4
1
33.3!x
+ 4.7x
+ [~x2 _ ~x3
2
3
18. (a) From y
1
+34.4!x
1
 4.7.10x
+ _1_x6
10
__
32 . 2!
12
 ...
1
+ 4.7.10.13x
1_x9
33 . 3!
+ _1_x12
34 . 4!
1 du
u dx
=   we obtain
Then dy / dx
= x2 + y2 becomes
;
1d u
dx2
+ u2
()2dudx
=x
+ u2
+ ~xl/2w
and
305
(~x2)
( dudx )2 '
13
]
oo.
_ ... ].
u" + x2u
Letting t
= xl/2
[x2w" Gx2)
+ 2w' Gx2)
= O.
~x2 we have
(2t  4 \t)
W(t)] = O
or
1/
= 1/4, so
(e) We have
y = _]:,du = _
u dx
xl/2w(t)
..!!:_xl/2w(t)
dx
= __ 1_
[xl/2dW dt + ~xl/2w]
dt dx
2
= __ 1_
[x3/2dW + _1_ w]
dt
2xl/2
xl/2w
xl/2w
=
1 [2x 2 di
dW]+ w =  2xw
1 [dW
2xw
4t di + w ] .
Now
4t
dw
so
306
+ C2( 
:/1/4(t)
 h/4(t))]
= el / e2 we have
19. Let
1
Y2 = x[ln(1
2
+ x)
 ln(1  x)]
so that
Y2,1[1
= x
2
1
+  1] + [ln(1
+ x)
l+x
lx
ln(1  x)]
and
"
1 [1
Y2 = 2"x  (1 + x)2
1[
+ (1 
1]
x)2
1[
1
1+ x
+ 2"
1]
1]
1 x
1[
+ 2"
1
1+ x
1]
x
+ 1
+ x)y~  2xy~ + 2m = o,
(1  x)(l
20. (a) By the binomial theorem we have
[1
+ (t2 
2xt
(b) Letting x
)] 1/2
= 1  2"1 (t 2 
2xt
+ '83 (t 2 
= 1 in (1 2xt+t2)1/2,
we have
+ t2)1/2
=
(1  2t
= (1  t)l
2xt
)2
= 1+ t
1t
+ t2 + t3 + ",
 1) t 2 + ....
(Itl < 1)
n=O
From part (a) in the text we have
00
:L Pn(l)tn
00
= (1  2t
+ t2)1/2 = :L e
n=O
n=O
Equating the coefficients of corresponding terms in the two series, we see that Pn(l) = 1.
Similarly, letting x = 1 we have
(1 + 2t
+ t2)1/2
= (1 + t)l
1
l+t
=
= 1 t
:L Pn(I)tn,
00
= :L(I)ntn
n=O
00
n=O
307
+ t2
 3t3
+ ".
(Itl < 1)
s 
(2
4e8tdt = _~est
s
= ;e
2.
o'i'{f(t)} =
Jo
1
= (  e
s
1 (
1
;  O ;e
6.
s2
f(
s>O
s>O
_~(e2S  1),
s
1 )  1(O e 8)
s2
= (~estsint
1 ( 1  e 8) , s>O
= s2
_ 21estcost)
s +1
s +1
+ 21estsint)
+ _1_e1<S/2)
82
+1
= __
82
1_e1<s/2,
+1
+1
11<
O
s> O
100
1</2
8> O
t)={O,O<t<l
t, i > 1
o'i'{f(t)} =
l tesi dt =
OO
8.
1
 ; ,
=O
7.
12 =
8
s  e
1 s)  ( 0;2  1) = 1( 1  3eS) + 2 ( 1  e S) ,
S
s2 S
S
s2
o'i'{f(t)} = r(sint)estdt
Jo
=
s) = ;e2
s  es21 s)  ( O
2 s  e
2
= ( e
5.
__1test __ 1)est
8
82
s2
O,
0< t < 1
f(t) = { 2t _ 2, t> 1
308
8> O
s>O
Exercses 7. 1
1  i,
:t'{f(t)}
10. f(t) =
12. :t'{f(t)}
13. :t'{f(t)}
0< t < 1
t> O
9. f(t) = { O
,
Jo
O,
O<t<a
e,
a < t < b;
O,
11=
82
~ es
82
8> O
82
5
= ( 00 e2t5estdt = e5 ( 00 e(s+2)tdt = __e_e(s+2)t
Jo
+ ~  ~,
Jo
+2
(00 ttestdt
Jo
Jo
4 8
1005= _e_,
e(4S)t)
(4  8)2
8> 2
+2
100
o
1
= (48)2'
14. :t'{f(t)}
1000 t2e2test
>4
di =
1000t2e(s+2)tdt
= ( __ 1_t2e(S+2)t _
8+2
15. :t'{f(t)}
1000 et(sin
= (
(8+1)
(5+1)2+1
2
te(s+2)t _
2
e(S+2)t) 100=
2
(8+2)2
(8+2)3
O (8+2)3'
t)estdt =
1000(sin t)e(s+l)tdt
e(s+l)tsint
 (8 + 1)2 + 1  82 + 28 + 2 '
1  ,e(1s)t
8
( (18)2+1
18
100
e(S+l)tcost)
(8+1)2+1
cos t
8> 1
1
(18)2+1
8 1
52 _ 25 + 2'
>1
309
e(1s)t sin t )
100
o
8> 2
Exercises 7. 1
17. Y{f(t)}
18. Y{f(t)}
= 10':;<) t(cost)estdt
=
st
[( 21
8 +
82  1
(82+1)2'
28). ?
(82+ 1)
(smt)e st]
28
(82+1)2'
28) 2 (cost)est(82+1)
(8t2+
8 +1
8>0
= 24~
20.
= 5!
Y{t5}
86
7 3
22. Y{7t + 3} = 2" + 8
8
2
2
6
3
23. Y {t + 6t  3} =  +   83 82 8
24. Y{4t
3!
2
3
1
25. Y{t3 + 3t2 + 3t + 1} = """"4 + 33" + 2" +8
8
8
8
26. Y{8t
t}
1
1
=  + 8 8 4
28. Y{t
121
+8 82
84
=  + 
33. Y {sinhkt}
+ 16t + 9} = 4
12t
8  k
CP
..L
2
+6t1}
{ 2t 2+e
= 2
8 + 25
8
2
8  k
= 2
1}_
{1
1
e 2t    1
2
2
2(8 2) 28
310
3!
84
2 6 1
83 82 8
= 812+
2t }=+1
82
3D. Ye
2
16 9
+  +83 82 8
9t
2
1
5
 e
+ 5} =   +83 8 + 9 8
34. Y {coshkt}
= 2k 2
(/){et sm
. h t} _
(/){ et et  et}
35. ..L..L2
.82
32. Y {cos5t + sin 2t}
36. Y{et
8>0
4
10
21. Y {4t  lO} =   82 8
27. Y {1+
00
= 1000 t(sint)estdt
= [( 2t
8 +1
19. Y{2t4}
2
8
1
8+2
+ 28 +4
Exercises
38. ~ {cos t}
=~
{1"2 + "21cos 2t } =
lIs
28
Q:
+ "2
82
+4
> O,
= 8 dt. Then
Q:
= r(1/2)
81/2
41. Identifying
For n = 1,
7. 1
> l.
f!_
V;
~{t}=~{1}=2'
For n = 2,
For n
= 3,
= t1/3.
311
F'(t)
Exercises 7. 1
Let
f (t)
2tet cos et
d
2
sin et . This function is not' of exponential order, but we can show that
dt
=:_
cos et 2 }
= 1000 est
(00
= s)o
d sin et 2) dt
dt
2
sin et 2 dt ]
= s :;t{sin et }.
est sin et dt
Since sin et2 is continuous and of exponential order, :;t {sin et2} exists, and therefore :;t {2tet2 cos et2}
exists.
43. The relation will be valid when s is greater than the maximum of C1 and
44. Since et is an increasing function and t2
C2.
> In M + ct for N! > O we have et2 > e1n M+ct = N!ect for
2
1
s  (a + ib)
1
(sa)+ib
(s  a)  ib (s  a)
s  a + ib
(sa)2+b2
+ ib
s a
= (s  a)2
+ i sin bt)}
+ b2 + i (s
b
 a)2
+ b2
bt} = (
sa
S 
r+a b')
and
:;t{eatsinbt}
= (
sa
~2
+ b2'
or
for all real numbers a and (3. Taking a = (3 = 1 we see that b = 2b, so b = O. Thus, f(x) = mx
will be a linear transformation when b
= O.
312
+b
Exercises
Exercises 7.2
3. ~
1 {1
48}
s2  s5
4!}
4
= ~ 1 {1s2  48
24 . S5 = t  2t
6. ~1{(S+32f}=~1{~+42.2+2.
s
s
s
23}=1+4t+2t2
s
7. ~1{12_~+_1_}=t_1+e2t
s
s
s 2
9.
~l
10.
{_l_}
4s + 1
~l
~l
{_l_}
_
5s  2 
~l
13.
~l
14.
~l
1 } = ~et/4
4
{~.
4 s + 1/4
1 }_ ~e2t/5
5
{~ .
5 s  2/5
4s } = ~l
4s2 + 1
1 }=
4s2 + 1
{~ .
~l
15. ~1{2s26}=~1{2_2s
s+9
s2 + 1/4
= cos ~t
1/2 } = ~ sin ~t
2 s2 + 1/4
2
2
s+9
7.2
2.~}=2COS3t2sin3t
s+9
313
Exercises
7.2
/2}
{s.
/2
S + 1 }
16. X' 1{ ~2
= X' 1 2 + 1! 2 = cosvr:
2t +  sin vr:
2t
s +
s + 2 v2 s + 2
2
17.
18.
19.
20.
1 } = X'1
X'1 { S + 1 } = X'1
s2 _ 4s
_1_}
{_~ . ~ + ~ . _1_} = _~ + ~e4t
4 s
4 s 4
4 4
X'1 {
S
} = X'1 {~ . _1_ + ~ . _1_} = ~et + ~e3t
s2 + 2s  3
4 s  1 4 s +3
4
4
X'1 {
X'1 {
s2 + 3s
sO.l
O)~s
)} = X'1 {(0.3). _1_
s+0.2.
sO.l
23. X'
26. X'
27.
s2 + S  20 
21. X'1 {(
22.
1{
X'1 {
{~ . ~ _ ~ .
= ~ _ ~e3t
3 s
3 s+3
3 3
)3)}
= X'1 {s2 ~ 3
(s2+4)(s+2)
28  4
1 {
(s2)(s3)(s6)
=X'
=X'
1 {
= 4 +
( 1{  1}
28. X'
s4  9
=X'
1 {1
..
6V3
s+0.2
)3 . s2~
1
2"s2s3+2"s6
1
s
1
2
's2+4+'s2+4's+2
} _ X'1 {
(s2 + s)(s2 + 1) 
+ (0.6). _1_}
0.3eo.lt + 0.6eO.2t
1}
1}
=2'e
2t
3t
e
+2'e
1
1 .
=cos2t+sm2t4e
2s  4 } _ X'1 {4
3
s
3}
s(s2 + 1)2 ~ + s + 1 + s2 + 1 + s2 + 1
)3
s2  3
6V3
V3}
s2  3
314
1.
r,;
=smhv3tsmv3t
6)3
1.
6/3
r:
6t
1 2t
Exercises
30.
68 + 3}
.:t'l { (82 + 1)(82 + 4)
= .:t'l
2}
{8
1
8
1
2 82 + 1 + 82 + 1  2 82 + 4  '2 . 82 + 4
1
8.:t{y}
 y(O)  .:t{y} = .
8
.:t'{y} =  + .
8
1
1
8
Thus
28.:t'{y}  2y(0)
= .:t'{y} = O.
8.:t{y}
 y(O) + 6.:t{y}
1
8_
4.
.:t'{y} = (84)(8+6)
2
+ 8+6
= 10' 84
19
19
1
+ 10' 8+6'
Thus
Y
4t
= 10e + 10e
6t
28
= 82 + 25 .
1118
= 13' 81
 1382+25
Thus
1
y = e
13
1
 cos5t+
13
315
5
sin5t.
13
55
+ 13' 82 +25'
7.2
Exercises 7.2
35. The Laplace transform of the differential equation is
S2 ~ {y}  sy( O)  y' (O) + 5 [s ~ {y}  y (O)] + 4 ~ {y} = O.
Solving for ~ {y} we obtain
s+5
~{y}
= s2
+ 5s + 4 = "3 s + 1 "3 s + 4
Thus
y = ~et _ ~e4t.
3
3
36. The Laplace transform of the differential equation is
s2 ~{y}
_ y(O)] = _6
s3
3_.
s+l
oL{y}
6
 (s  3)(s2  4s)  (s
5 1
2
= _.   2 s
s  3
3
+ 1)(s2  4s)
s5
s2  4s
+=
3
1
11
1
+ .
o
5 s + 1 10 s  4
 _. 
Thus
y
5
2
3
5
= __ 2e3t __ et
11
+ _e4t
10
 sy(O) +~{y}
~2'
we obtain
~
{y}
+~=~+_2
(s2+1)(s2+2)
s2+1
2_
s2+1
s2+1
s2+2'
Thus
y = lOcos t + 2 sin t 
h sin ht.
+ 9~{y}
= .
s 1
1
= (81)(s2+9)
11111
= 10' sl
 10' s2+9
Thus
y
1 tI.
= e
 sm3t
10
30
316
1
 cos3t.
10
10'
s
s2 +9'
7.2
Exercises
39. The Laplace transform of the differential equation is
2 [s3.:t'{y}  s2(0)  sy'(O)  y"(O)] +3[s2 .:t'{y}sy(0)y'(0)]3[s
.:t'{y}y(0)]2
.:t'{y} = _1_.
s+l
.:t'
{y}
2s + 3
(s+1)(s1)(2s+1)(s+2)
= ~ "_1_ + ~
2s+1
_1 __
18s1
+ ~ _1_
9s+1/2
9s+2"
Thus
1
5 __ 8et/2 + _e2t
1
Y = _et
+ _et
2
18
s3 .:t'{y}  s2(0)  sy' (O) y" (O)+ 2[s2 .:t'{y}  sy(O)  y' (0)] [s .:t'{y}  y(O)] 2 .:t'{y} = +9
s
"
.:t'{}s2+12
y  (s  1)(s + 1)(s + 2)(s2 + 9)
13
= 60
13
16
s  1  20 s + 1 + 39 s + 2 + 130 s2 + 9  65 s2 + 9 .
Thus
13 t
Y = e
60
= 1 and g(t) = {
13 t
 e
20
16 2t
3
+ e
+ cos3t
39
130
1 "
 sm3t.
65
t > O, t =f. 1
. Then .:t'{f(t)} =.:t'{g(t)}
O, t = 1
1,
.:t'{}
.:t' { }
s +3
s +3
y +
y = (s + 3)2 + 4 = s2 + 6s + 13 .
.:t'{ }_
Y
s+3
(s+1)(s2+6s+13)
4 .s +1  4
=4'
1 (S+3
s+14'
s+l
s2+6s+13
2)
(s + 3)2 + 4  (s + 3)2 + 4 .
Thus
1
yo(t)=.:t'l{
s5
}=.:t'l{~.~_~._l_}=~_~t,
s2  4s
4 s
4 s  4
317
Exercises 7.2
and the zerostate response is
Yl (t) =:L
1
{6
=:Ll{27.
3}
_1
20 3  4
2_ + ~. ~ _ ~. _1_}
3 3 4 3 5 3+ 1
Exercises 7.3
1.
:L {telOt} =
2. :L {6t}
te
3.
1
(3  10)2
1
= (3 + 6)2
:L{t3e2t} =
4. :L{t10e7t}
5
3!
(3 + 2)4
lO!
(3 + 7)11
:L{t(et+e2t)2}=:L{te2t+2te3t+tt}=
1
(32)2
2
6.
2
(33)2
2
2t
(3_13)2+9
3+2
= (3+2)2+16
8.
:L { e
9.
:L{(1  et + 3e4t) cos 5t} =:L {cos 5t  et cos5t + 3e4t cos 5t}
cos4t
3
 32+25
11.
1
(34)2
7. :L{etsin3t}=
10.
t) } =:L {3tge
( { 3t(
.
:L e 9  4t + 10 sm"2
:Ll{
31
(31)2+25
 4te
3t
3(3+4)
+ ,':,,(3+4)2+25
t}
3t .
9
4
5
+ lOe sm"2 = 3 _ 3  (3 _ 3)2 + (3 _ 3)2 + 1/4
1
} _ :Ll{~
2
} _ ~t2e2t
(3+2)3
2 (3+2)3
 2
318
7.3
Exercises
{(S ~ 1)4}
= ~l
(S ~!1)4}
= ~t3et
12.
~l
13.
~l {2s  6s1 + 10} = ~l {(s  3)\ + 12}= e3t sin t
14.
15.
16.
~l{s2 +2s+5
}=~1{2 (s +(s+3)
_~
5
}=2e3tcos5t~e3tsin5t
6s + 34
3)2 + 52 5 (s + 3)2 + 52
5
17.
18.
19.
20.
{~
(s +22) 22
12 2}=e2tcost2e2tsint
(s + 2) + 1
(s + 2) + 1
~l {(s: 1)2} =~l {s(:: 0;}=~l {s ~ 1  (s: 1)2} = et  u:'
~l {(s2)2
5s } = ~1 {5(S  2) + lO} =~1 {_5_ +
10 } = 5e2t + 10te2t
(s2)2
s2
(s2)2
2s  1}
{5;  s21  S +5 1  (s +4 1)2  2"3 (s +2}1)3 = 5t5e 4te
~
s2(s + 1)3 = ~
1 {
~1
t
1
{(S + 1)2} =
(s+2)4
~1
}=
1
_
2
+~
3!
(s+2)2
(s+2)3
6(s+2)4
u:? _ t2e2t
+ ~t3e2t
6
s~{y}
Solving for ~{y}
we obtain ~{y}
= (s
 y(O) + 4~{y}
1
+ 4)2 +
y
= 4'
s+
+ 4' Thus
= te4t + 2e4t.
s ~{y}  ~{y}
Solving for ~{y}
1
=;
+ (s _ 1)2 .
we obtain
~{y}
s(s  1) + (8  1)3 =
Thus
319
;
+ s1 + (81=)3
.
t
2 t
2"t e
Exercises 7.3
23. The Laplace transform of the differential equation is
 y(O)] + :t'{y} = O.
we obtain
s
+3
we obtain
(s ~ 2)4'
:t' y} = 1 + s2
{
s2(s  3)2
= 2 ~+ ~~ _ 2 _1_ + 10
27 s
9 s2
27 s  3
1
9 (s  3)2 .
Thus
2
y=27+9t27e
2 3t
10 3t
+gte
= 4"'
s
:t' y = s5  4s4 + 6 = ~ ~ + ~~ + ~!
+ ~ ~ + ~ _1 __
{}
s4(S  2)2
4 s
8 s2
4 s3
4 s4
4 s  2
13
8
Thus
Y=
3 2
1 2t
+ st + t + t + e 
13 2t
ste
320
(s  2)2 .
Exercises
7.3
Thus
y
3
.
=  2e3t
SIl1 2t.
2[82.:t'{y}
= O.
{y}
48+40
282+208+51
28+20
(8+5)2+1/2
2(8+5)
(8+5)2+1/2
10
(8+5)2+1/2'
Thus
8 
(8 _
1) 2 + 1 .
{y}
=~~_~
1
8(8228+2)
81
+~
282(81)2+12(81)2+1'
Thus
1 1t
1t.
y =   e cos t + e sm t.
2 2
2
30. The Laplace transform of the differential equation is
= ~ + 12,
.:t'{y}
2. ~
482 + 8 + 1 =
+ ~ _.!:_ + 78/25 + 109/25
82(82  28 + 5)
25 8 5 82
82  28 + 5
7 1 1 1
7
8 1
25858225(81)2+2225(81)2+22'
=   +   
51
+  ,...".....",
Thus
7
Y=25
+ t1
5
7 t
 e cos 2t
25
51 t .
+ e
sm 2t.
25
31. Taking the Laplace transform of both sides of the differential equation and letting e = y(O) we
321
Exercses 7.3
obtain
s2~{y}
=O
=O
~{
es + 2e + 2
}_
y
es
(s
=e
+2
+ 1)2
2e
+ 1)2 +
(s
s +1 1
2e + 2
+ ..,.=
(s+1)2
(s+l)2
e+2
+ ..,.,'"
(s+1)2'
= s+l
Therefore,
2. Then 2
ee1
and e
e  l. Thus
y' (O) we
obtain
 y'(O) + 8s~{y}
s2~{y}
 e+8s~{y}
+ 20~{y}
=O
+20~{y}
=O
(i+8s+20)~{y}=e
~{
} _
y  s2 + 8s + 20  (s + 4)2 + 4 .
Therefore,
y (t )
<..01 {
oL
s+4
e)2
+4
1l'
= ce 4t
and e
sin 2 t.
differential equation is homogeneous and both boundary conditions are O, we can see immediately
that y(t)
Exercises
= O.
+ 7x' + 16x
82:t'{X} +
~8
+ 78:t'{X} + 21 + 16:t'{x} = O.
2
82 + 78 + 16 
8 + 7/2
7V15
2 (8 + 7/2)2 + (V15/2)2
10
Vl5/2
(8 + 7/2)2 + (V15/2)2 .
Thus
3 7t/2 cost
Vl5  e
7Vl5
x = e
2
2
10
7t/2
. Vl5
smt.
2
d2q
dq
dt2 + 20 dt + 200q = 150,
q(O) = q'(O) = O.
:t'{ } _
150
q  8(82 + 208 + 200)
3 1
4' ~  4'
8 + 10
3
10
(8 + 10)2 + 102  4' (8 + 10) 2 + 102 .
Thus
3 lOt
3 lOt .
q(t) = 3  e
cos lOt  e
sm 10t
444
and
d2q
dq
2
Eo
dt2 + 2'\ dt + w q = L'
q(O) = q'(O) = O.
or
(82 + 2'\8
+ w2):t'{q}
= ~o ~.
:t'{q} = Eo (1/W2
L
7.3
+ w2
323
Lw2
+ 2,\
82 + 2'\8
+ w2
) .
Exercises
For A
7.3
1
S+A
A)
,,;t'{q} = EaC ( ~  (s + A)2 _ (A2 _ w2)  (s + A)2 _ (A2 _ w2)
Thus for A
> w,
q(t) = EaC(l
For A
= w,
VA2
w2
sinh
VA2 
w2 t) .
so
A)
s+
(S+A)2+(W2A2)
= (s + A)2
s2 + 2A + w2
{q}
Thus for A
w2 t 
 (S+A)2+(W2A2)
< w,
,,;t'
VA2 
= (s
1
( ~
,,;t'{q}=Ea
For A
eAt cosh
Thus for A
and
= Ea (l/A
Ea
1
L S(S+A)2
l/A
S+A
l/A ) = Ea (~ __ 1_ _
A )
(S+A)2
LA2 s S+A
(S+A)2
= w,
dq
R dt + C q = Ece
kt
= O.
,q(O)
+ C ,,;t'{ q}
Ea s + k .
EaC
Ea/ R
,,;t'{q} = (s + k)(RCs + 1) = (s + k)(s + 1/ RC) .
When 1/ RC =1 k we have by partial fractions
::t { }
s + 1/RC
s +k
Ea
R
1
1/RC 
Thus
q(
When 1/ RC
=k
t)
EaC
(kt
ti Re)
kRC e  e
.
1_
we have
,,;t'{q} =
Ea
(s + k)2 .
324
(_1_ _
k
s +k
s + 1/RC
).
Exercises
7.3
Thus
39 .
= .:t {e(t2)
OU(t _ 2)}
2s
= _e _
s+1
e2s
2e2s
.:t{tOU (t  2)} = .:t {(t  2)OU(t  2) +20U (t  2)} = 2 +s
es
40. .:t { (3t + 1)OU(t  1)} = 3 .:t { (t  1)OU(t  1)} + 4.:t { OU(t  1)} = 2 +
4es
45 . .:t1 L~:s1}
46 .
= cos2 (t 
es
s(8+1)
)}
s81
49. (c)
55.
56.
57.
se7rS
.:t { cos 2 (t  71') OU(t  71')} = 82 + 4
.:t1 {8:;:S~2}
47 .. .:t1 {
= .:t1
= .:t1
%) OU(t _ %)
{eS _ eS}
8
8+1
= OU(t
_ 1) _
50. (e)
51. (f)
e(tl)
OU(t  1)
= _ OU(t _
2)  (t  2) OU(t  2) + 2ou (t  2)
53. (a)
52. (b)
54. (d)
1 e4s
e5s
.:t {1  OU(t  4) + OU(t  5)} =   +8
S
s
.:t {t2 OU(t  1) } = .:t { ((t  1)2+ 2t  1]OU(t  1)} = .:t { ((t
325
Exercises
7.3
2
2
= ( ++s3 s2
1)_s
e
s
1 e2s
s2 ~s
59. 2{ttOU(t2)}=2{t(t2)OU(t2)2OU(t2)}=
60. 2{sintsint
OU(t27f)} = 2{sintsin(t27f)OU(t27f)}
eas
ebs
= s2+1
es
e2s
2e2s
e27rs
 s2+1
e3s
es
y(O) +2{y}
5
s
= _es.
2{ } = 5eS
y
s(s+1)
= 5es [~ __
1_]
s+1
Thus
y = 5OU(t 1)  5e(tl)OU(t 1).
64. The Laplace transforrn of the differential equation is
s 2{y}  y(O) +2 {y}
Solving for 2{y}
= ~  ~es.
s
we obtain
1
2{y}
= s(s
+ 1) 
2es
1
1
8 [1
1]
s(s + 1) = ~  s + 1  2e
~ s +1
Thus
1
s
? 
s +1
es2.
s
1
s2(s+2)
8
e
s+1
s2(s+1)
11 1 1
1
=4~+2s2+4s+2e
326
s
[1 1 1 1
1
4~+2s24s+2'
1 ]
Exercses
7.3
Thus
y = ~ + ~t + ~e2t  [~+ ~(t  1)  ~e2(tl)]q(t
4 2
4
4 2
4
 1).
+ 4.:t'{y}
s
e .
5
s
.:t'{y} = 5(52+4) e
1
5(52+4)
11
=4~4
5
52+4
2'
2
s [1 1
1 5
52+4 e
4~452+4'
Thus
y = ~  ~ cos 2t  ~ sin 2t  [~  ~ cos 2(t  1)] q(t  1).
4 4
2
4 4
67. The Laplace transform of the differential equation is
52.:t'{y}
= e27rS~1
5
Thus
y = cos 2t + [~sin (t  27r) ~ sin 2(t  27r)]q (t  27r).
68. The Laplace transform of the differental equation is
52.:t'{y}
es
= .
5
= es
[~ ~  ~ 5 ~ 2 + ~ 5 ~
3] 
5~ 2 + 5~
_ 1)
+ e3t
3.
Thus
y = [~_ ~e2(tl)
6 2
+ ~e3(tl)]q(t
3
_ e2t.
= _e
27r5
e _.
5
.:t'{y} = = [~__
5
_1_.
52 + 1
327
Exercses 7.3
Thus
y
s2 ~{y}
_ sy(O) _ y'(O)
+ 4[s~{y}
_ y(O)]
+ 3~{y}
28
48
e _
s
68
+ _e
++".
2s+1
6s+3
3s
2s+1
+ ~ _1_] + e68
6s+3
+ ~ _1_]
[~~ _ ~ _1_
3s
2s+1
6s+3
6s+3
Thus
1 1 t
y = 3"  2e
 [~_
3
+ "61e 3t
~e(t4)
+ ~e3(t4)]OU(t
[1
1 (t2)
3"  2e
_ 4)
+ "61e 3(t2)]
(
OUt  2)
+ [~3
+ f(t).
+ ~e3(t6)]OU(t
~e(t6)
2
x(O) = O, x'(O)
= O.
20t,
f(t)
Also, since
f(t)
 6).
+ 100 we can
={
O,
O:::; t
<5
t >: 5
write
 5).
20
_
20
e58 _
100
e58
 s2(s2 + 16) s2(s2 + 16)
s(s2 + 16)
=(~.2__~.
4 s2
16
s2
+ 16
)(1_e58)_(25.~_25.
.5
s2
+ 16
)e58.
Thus
x(t) = ~t  156sin4t  [~(t  5)  156sin4(t  5)] OU(t  5)  [~5  245cos4(t
5
25
 5)] OU(t  5)
Exercises 7.3
72. Recall from Section 5.1 that mx"
kx + f(t).
Now m
< 27T
O:::;t
sin i,
f(t)
={
O,
t 2: 27T
82X{X}
+ 16x{x}
= 21
1
 21e27rS.
8
VJ{ X }
oL.
= 1/15
82 + 16
e 27rS
 (82+16)(82+1)
 (82+16)(82+1)
Thus
15
60
15
O:::;t
< 27T
t 2: 27T.
O,
2.5 dq
dt
+ 12.5q
= 5OU(t
 3).
8X{q}
+ 5x{q}
=
e3s.
x{q}
e3s = (~ . ~ _ ~ .
8(8+5)
5 8
_1_)
5 8+5
Thus
q(t) = ~OU(t  3) 5
~e5(t3)
OU(t  3).
329
30etOU(t  1.5).
e3s.
Exercises 7.3
The Laplace transform of this equation is
3
3e1.5
8.;t'{q}  qO +.;t' {q} = 8 _ 1  8 _ 1.5 e1.5s.
Solving for .;t'{q} we obtain
.;t'{q} = (qO _ ~) . _1_ + ~. _1_3e1.5
2
8+ 1 2 8 1
(2/5 +~)
e1.55.
8 + 1 8  1.5
Thus
(t _ 37r)OU(t_
37r)
22'
= sin z } cos
~
The Laplace transform of this equation is
di + 10i
1
8.;t'{i} + 10.;t'{i}
i(O) = O.
8e31rs/2
= 82 + 1 + 82 + 1 .
1
(82 + 1)(8 + 10)
__ 1_ (_1
 101 8 + 10
e31rs/2
8_ + ~)
+ _1_ (~
+ ~
+ _1_)
e31rs/2
82 + 1 82 + 1
101 8 + 10 82 + 1 82 + 1
.
Thus
1
0'2f
+~""'"~..__.=::::,._~._.__;:",.,_~::::"'~~~..__
1
3~
t
6
0.2
The mximum
0.1 at 4.7.
76.
330
q(O) = O
Exercises
50 ~;
+ 100q = Eo[VU(t 
q(O) = O.
1) VU(t  3)],
+ 100:t{q} = Eo
50s:t{q}
Ges
 ~e3S).
:t
{q}
_ Eo [ es
50 s(s+2)
e3s
s(s+2)
] _ Eo [~ (~ __ 1 ) es _ ~ (~ __ 1 ) e3S]
 50 2 s s+2
2 s s+2
.
Thus
(b)
= y'(O) = O we obtain
(1 _ eLs/2)
:t{y} = 2 +
s3
C2
s4
+ Wo ~
El
s5
(1 _ eLs/2)
so that
1
= ClX
y(X)
To find Cl and
C2
2 + C2X
1 3 + 1 Wo [4x
6
24 El
(x
 
we compute
y
"(x) = Cl
( x "2L)2
1/(
u x 
and
y'" (x) = C2 +
~~ [x  (x 
~)vu ( x 
331
~) ] .
L)]
"2
7.3
Exercises 7.3
Then y" (L)
=;
v" (L)
wo (L)
Solving for
and
Cl
we obtain
C2
Cl
"2
+ El
C2
wOL2 = O
= Cl + C2L + 3S El
= C2
+ '2 El
1 woL
~wOL2/ El and
C2
=
= O.
iwoL / E l. Thus
wo (1 2 2
1 3
1 4
1 (
y(x) = El 16L x  12Lx + 24 x  24 x
"2L)4
(
L))
/Jx  "2
d4y
Letting y"(O)
= Cl
e2Ls/3).
Cl
and
C2
1 3 + 24
1 El
wo [( x  '3
L)4 /Jx
(  '3
L)  ( x  '3
2L)4 /Jx
(  '3
2L)]
+ 6C2X
we compute
and
 (L)2]

Cl
Wo [2L
C2
C2
we obtain
Cl
Wo (112L 2 x 2  lS1Lx
y(x) = El
=
3
+ El
twoL2/EI
wOL2
+ C2L + 1 6 El
L]
'3 '3
and
C2
=O
1 woL
C2
+ '3 El = O.
~woL/EI.
Thus
1 [( x  '3
L)4 /Jx
(  '3
L)  ( x  3
2L)4 /Jx
(  '3
2L)])
+ 24
332
Exercises
79. The differential equation is
El d4y = 2wo
dx4
L
[!:_
x +(x  !:) a (x _ !:)] .
2
2
2
= Cl
Letting y"(O)
2s
s2
s2
= C2 we have
and y"'(O)
UJ{ } _++Cl
C2
2wQ [ L
1
1 LS/2]
+e
s3
s4
ElL
2s5
s6
s6
""Y
so that
y(X) = ~Clx2
2
=
To find
Cl
and
120
!:)
2
_1_ (X 
120
L)
+ ( X  "2
5 (i'1
=u.
(X _
lJij
!:)]
2
"2L)] .
X
we compute
C2
y"(x) = Cl
o (X  ~)]
and
Y /1' (X)=C2+60ElL WQ
Then y" (L) = u" (L)
ci
= Oyields the
[60Lx60x
Cl
and
y(x)
Solving for
C2
we obtain
wQL2
= 48ElX
2
C2
+ 60El L[60L
Cl
= woL2/24El
woL
24El
+60
L)
x"2
2 o
( x"2'
L)]
system
[3
WQ
WQ
+ 60ElL
3] =
 60L
and
[5L
2X
C2
5woL2
Cl
+ C2L + 24El
woL
+ 15L ] =
C2
+ 4El
= woL/4El.
4  X5'(+
X
=O
= O.
Thus
L)5
"2
(
L)]
o X  "2
= wo[l  o (x  L/2)J.
Letting y"(O)
= ci
3. +
s3
C2
s4
+ Wo ~
El s5
333
(1 _ eLs/2).
(1 eLs/2)
7.3
Exercises 7.3
so that
1 2 + C2X
1 3 + 1 Wo [4X y(X) = C1X
2
6
24 El
To find
Cl
and
C2
= y"(L)
1
clL
2
ci
(  "2
L)] .
+ C2X+ "21WO[2
El X  ( X "2L)2 llUx
2 +C2L
1
3 +1 Wo [4L
6
24 El
C
Solving for
we compute
y " (x) = Cl
Then y(L)
(X
and
C2
we obtain
4]
 (L)
2
1
2
= clL
Wo [2L "2(L)
+ C2L + "21 El
Cl
1~8
woL2/EI and
2 +C2L
1
3 +L5wo
2]
C2
128EI
4 =0
3wo L 2 = O.
= ci + C2L + 8EI
= lS278
Wo (9
2 2
19 3
1 4
1 (
y(x) = El 256L x  256Lx + 24x  24 x
wo~/EI.
"2L)4
r711(
'u,
Thus
"2L)) .
81. In order to apply Theorem 7.7 we need the function to have the form f(ta)llU(ta).
To accomplish
+ 1 = 2(t 
1 + 1) + 1 = 2 (t  1) + 3
= 1/(8 
(8  ki)2
(8+ki)2
82_k2
.2k8
=
+ Z =,,,;:(82 + k2)2
(82 + k2)2
(82 + k2)2 .
:t{tcoskt}
= (82
k2
+ k2)2
and
2k8
:t{t sin kt} = (82 + k2)2'
82 :t {x} + w2 :t {x} =
Solving for :t{x}
we obtain :t{x}
2 8
8
+w
2'
= (1/2w)tsinwt.
Exercises
Exercises 7.4
8.::tt {
9.
ie
t}
=s3(s1)2
sl
(s+1)[(s1)2+1J
::t{et*etcost}=
(s2)(s2+1)
10. ::t{e2t*sint}
= ~::t{et} = S(s~ 1)
12. ::t{t
COSTdT} = ~::t{cost}
t
14. ::t{l
TSinTdT}
= S(S2S+ 1)
s2 ~ 1
1
t
1
S + 1
= ~ ::t{e
cost} = ~ (s+1)2+1
= ~ ::t{tsint}
= ::t{t}::t{et}
1(
=~ 
7.4
dI)
+1
s(s2+2s+2)
1  2s
_
2
ds s2 + 1 = ~ (s2 + 1)2  (s2 + 1)2
S2(Sl_1)
335
Exercises
7.4
=.:t'{sint}::t{cost}
Jo
17.
::t{t
r Sinrdr}
= !!:_::t{ rt Sinrdr}
Jo
=  :S .:t'{t
(b) ::t1{
(e) ::t1{
s(s  1)
21
s3(s  1)
reTdr}
{l/(S
} = ::t1
s (s  1)
Jo
ds
dd
(~2_+11_) = s2(s2+1)
3s2 + 1 2
rt(eT_1)dr=et_t_1
Jo
=::t_1{1/S2(S1)}
s
S ss
=  :S
}=::t_1{1/S(S1)}=
s
}
s
(s2+1)
::t1{F(s)G(s)}
=f
(t(eTr_1)dr=et_~t2_t_1
Jo
* g, so identify
2k3
F(s) = (s2 + k2)2
and
4s
G(s) = s2 + k2 .
Then
f(t)
= sinkt
 ktcoskt
g(t) = 4coskt
and
so
::t1 {(S2S:3;2)2 } = ::t1 {F(s)G(s)}
=f
*9 =4
f(r)g(t
 r)dt
 r)dr.
::t
l{
(s2Sk+3S}.
k2)3 . = tsm kt  kt cos kt.
F'(s).
336
Sk3s
(s2 + k2)3'
Exercses 7.4
22.
:f{f(t)}
23.
:f{f(t)}
25.
:f{f(t)}
26. :f{f(t)}
12
(a estdt
1  e as Jo
1
= 8 (1 +1eas )
io
1f
1  e1fS o
1 2
1 e 1fS
+1
82
e1fsj2
e1fsj2
Jo
1fS
Thus
1
Y (t) =   e t
2
=
1.1
l.
l.
+ l' cost
1t
e
1
 tcost
2
+ l.r sm t.
2
822{y}
337
= ~9'
8
Exercises 7.4
Letting y(O)
= 2 and
y'(O)
;t
= 5 and
we obtain
{y}
(52+9)2
+ _5_
+
5
52+9
(52+9)2'
52+9
Thus
5 ,
1,
52;t{y}
= 21,
5 +1
or
= 1+ 25
(52+ 16);t{y}
+ 16
e7fS;t{cos4(t+7f)}
= 1 + 52 + 16  e7fS ;t{cos4t}
= 1+
52 + 16
e7fS
52 + 16
Thus
;t{ }
1
5
y =52+16+(52+16)2(52+16)2e
7fS
and
4
8
8
32. The Laplace transform of the differential equation is
52 ;t {y}  5y(0)  y' (O) +;t {y} =;t { 1 Ilfl (t  ~) + sin t OU( t  ~) }
or
(52 + 1) ;t{y}
= 5+ ~  ~
52 + 1
338
Exercses 7.4
Thus
.,'i'{ }
s
s2
+ 1 + s(s2 + 1)
S
= s2
+ 1 + ~
1 (1~ 
= ~
s
s2
+1
S)
s2
1
 s(s2
+1
+ 1)
(1~ 
1rs/2
e1fs/2
S)
s2
+1
s
(s2
(s2
1fs/2
+ 1)2 e
+ 1)2
+
e1fs/2
S
(s2
+ 1)2 e
1fs/2
1fs/2
and
(t  ~) ] VU (t  ~) + ~ (t  ~) sin (t  ~) VU (t  ~)
(1  sin t) VU (t  ~) + ~ (t  ~) cos t VU (t  ~) .
y = 1  [1  cos
= 1
s .,'i'{y}
2s
+ .,'i'{y}
= (s2
+ 1) + (s2 + 1) 2 .
2s
50
Thus
~{y}
(s2
50
34. (a)
y
1
1
(b)
y
4
2
1
2
4
339
Exercises 7.4
35. The Laplace transform of the given .equation is
~{f}
Solving for ~{f}
+~{t}~{f}
1
= 2.
we obtain ~{f}
Thus, f(t)
+1
= ~{t}.
= sin t.
= ~{2t}
~{f}
 4~{sint}~{f}.
= 2s2 + 2
= ~ 2_
s2(s2+5)
{f}
5s2
+ _8_
V5
5V5s2+5
Thus
28.
= t +
f (t)
~ sin Vs t.
5v5
~{
}_
_ ~ _1_ + ~
s2
1
+ ~ 2 _ ~ _1_
4 (s  1)2 4 (s  1)3 88+ 1 .
f  (s  1)3(s + 1)  8 s  1
Thus
f(t)
1 t 3 t 12 t
= e
+ te + t e el_t
8
~{f}
Solving for ~{f}
+ 2~{cos
t}~{f}
we obtain
~{
} _ 4s2 + s + 5 __ 4_ _
7
+4
2
(s + 1)3  S + 1 (s + 1)2
(s + 1)3.
f 
Thus
~{f}
Solving for ~{f}
we obtain ~{f}
+~{1}~{f}
1
= .
s+l
= ~{1}.
340
Exercses 7.4
40. The Laplace transform of the given equation is
.:t'{f}  _
82
+1
+ _1
82 + 1 '
Thus
+ 1)
.:t'{f} =
82(8
84 +
16 =
83
84 + 16
82
84 + 16
'
Thus
f (t)
= cos
J2 t cosh .J2 t +
2v2
.:t'{f}
1 1
= 2T = 2 82
1 31
12 84
'
Thus
1
2
1 3
.
12
f(t) = t  t
43. The Laplace transform of the given equation is
8 .:t' {y}
.:t'
= 83  82
{y}
+8 =
8(82+1)2
_1_
+~
82+1
28
2(82+1)2'
Thus
.
1,
Y = sin t  2 t sin t.
44. The Laplace transform of the given equation is
.:t'{1}.
Exercises 7.4
we obtain ~{y}
= (
8+3
+ 3i + 0.05 lo
20
10
or
si
.5 3
Jo
10
20
where i(O) = O. The Laplace
.
200
8~{2}  y(O) + 30~{i} + ~{i}
1000
= _(eS
3 O
_ e2S).
= 1000es
1000e2s
82+308+200
~)(eS
8+108+20
_ e2s).
= (~
Thus
i(t) = 100(e10(tl)  e20(tl)) )L(t  1)  100(e10(t2)  e20(t2))
i
)L(t  2).
lo i(T)dT
t
= 100[t  (t l))L(t
r.,
1)]
1.5
or
d'
t
__:.+ 200i + 10,000 { i(T)dT
8t
Jo
0.5
1.5
Thus
i(t) = 2  2e100t  200te100t
342
Exercises
di
dt
where i(O)
7.4
i
1
+ i = E(t),
equation is
s:t{i}
0,5
+ :t{i}
= ~{E(t)},
1
1 es
= s (1 + eS )'
Thus
(s
+ 1) :t{i}
1 e s
s + e S)
= (1
and
1 e s
~ = s(s + 1)(1 +
:t{"}
= (~ __
s+ 1
s
=
eS)
1 e s
s(s+l)l+es
( 1s  s+l1)
(1  2e s
_ es
+ 2e 2s
+ e2s
_ e3s
 2e 3s
+ e4s
+ 2e 4s
_ " ,)
 , , , ),
Therefore
i(t) = [1 2OU(t 1)
 [et
+ 2OU(t  2) 
+ 2e(tl)
+ 2e(t3)
2OU(t '3)
OU(t  1)  2e(t2)
OU(t 3)  2e(t4)
+ 2OU(t  4) " ,]
OU(t  2)
OU(t  4)
00
= 1 et +2 2:::(lt(1
e(tn))OU(t
n=l
343
 n).
+ ... ]
Exercises 7.4
di .
dt + 2 = E(t),
1
0.5
+ :t{i}
s:t{i}
:t{E(t)}.
1
:t{E(t)}
= 1(1
   1)
= 1  1 . 1
s s eS  1
s2 s eS  1
Thus
(s + 1) :t{i}
and
:t{"}
2
= s2(s + 1)
Therefore
1 1
( +s2 S
1 1 1
 s2 s eS  1
=
1
1
s(s+1)es1
1)  (1
 1) (S
e
S+1
s s +1
(t  1+ et) 
+e
2s
+e
38 +e 48
 3)  (1 e(t4))r7U(t
 .... )
 2)
 4)  ...
00
L (1 e(tn))r7U
(t  n).
n=l
Using the initial conditions x(O) = x'(O) = Oand taking the Laplace transforrn we obtain
= 20 (1 _ e7rs)__ l_
1 + e7rS
20
344
= tt .
Exercses 7.4
Then
.:t'{ (t)}
X
. 20
40
~ ( l)n n7rS
= 5(52+ 25 + 10) + 5(52+ 25+ 10) ~l e
2
=~
25 + 4
~ ( )n [4
45+ 8 ] n7rS
52+25+1O+~
1
~ 52+25+10 e
=~_2(5+1)+2+4f.
5 (5+1)2+9
(_l)n[~_
(5+1)+1]en7rS
(5+1)2+9
11=1
and
+ 4~(l)n[l
 ~etsn3t)
e(tn7r)
cos3(t  mr)
3
+ 2x' + x
= 5f(t),
Using the initial conditions x(O) = x'(O) = O and taking the Laplace transform, we obtain
(52+ 25 + 1).:t'{x(t)} = ~
5
1 + e7rS
Then
.:t'{x(t)} =
5
f (_l)"en7rs
5(5 + 1)2 n=O
= 5 f (1)" (~ __ 1_ _
n=O
5+ 1
1 ) en7rs
(5 + 1)2
and
00
x(t) = 5
L (1)"(1
e(tn7r) 
n=O
345
(t  nn)e(tn7r))
lJU(t nn).
Exercises
7.4
5
51. f(t)
= ~
.,;t'l{ ~(ln(s
ds
l_} = _~t (
+1
t 
et)
52. By definition, t *q(t  a) = IJ(t  7)q(7  a) dr . We consider separately the cases when a
and when a
(t (t _ 7) q (7 _ a) dr
Jo
When a
> t, then q (7  a)
l;
7 < t < a and
= O since
(t  7) q (7  a) dr = O.
Therefore
hq(ta)
> a = ~(ta)2q(ta).
t <a 2
{i(ta)2,
O,
53. First method: By the definition of the Laplace transform and integration by parts we have
{l f (7)
.,;t'
dr }
dt
s Jo
s Jo
dt = F(s) .
Jo
Second Method: Let g(t) = Id f(7) dr ; then g'(t) = f(t) and g(O)
= O. By Theorem
and
.,;t'{f(t)} = s.,;t'
.,;t'{l
f(7) d7}
{la
f(7) d7} ,
= ~.,;t'{f(t)}
346
F;S) .
7.8
<t
Exercses
54. Let u
7.4
= t  T so that du = dr and
* 9 = 10t f(T)g(t
 T) dr = 
iD f(t
 u)g(u)du = 9 * f.
55. (a) Using Theorem 7.8, the Laplace transforrn of the differential equation is
d
= _
dx
[s2y] + sY +  [sY] + nY
ds
=:
(s  s2)
y)  2sY + sY + s(:s y) + y + nY
(!
y) + (1 + n  s)Y = O.
dY = 1 + ns ds = (_n __
y
s2  S
S  1
In Y = n ln(s ~ 1)  (1 + n) In s
Y
_
 CI
1 + n) ds
+c
(s  l)n
sl+n
Since the differential equation is homogeneous, any constant multiple of a solution will still
be a solution, so for convenience we take
el
n=O:
n = 1:
n = 2:
n
= 3:
n = 4:'
= O for k = 1, 2, 3, ... ,
347
Exercises 7.4
Now, by the first translation theorem,
where Y
=::t {Ln(t)}.
Thus
56. (a) The output for the first three lines of the prograrn are
+ 8 Y + 6(2 + 8Y)
==
28
(1 + 82)2
>
a solution involving complex coefficients of eit and e:". To get the solution in more standard
form write the last line as two lines:
euler = {E~(I t)> Cos[t]+1
InverseLaplaceTransform
Sin[t]}
247)
y(t) = ( 250 + 5t e3t + 250 (13 cos t  15t cos t  9 SIn t + 20t sin t) .
(b) The solution is
348
Exercises
7.5
Exercises 7.5
.:t'{y} = ~3e2S
8
so that
y
= e3(t2)q,(t  2).
es
8+1
8+1
+ e(tl)q, (t 
1).
.:t'{ } = ~
y
82
+ 16
e2Tfs
so that
y
.
41 sm4(t
 27f)q,(t  27f).
+1
349
Exercises 7.5
so that
y
+1
82
+1
+ e471'S)
so that
y
.s:'{y} =
so that
y
1 2t + [1
1 2(tl)]IilI(
= 1  e
  e
'u t  1).
2
.s:' y =
{}
8+ 1 +
1
e2s = ~_1 __
82(82)
8(82)
482
~ ~ _ ~ ~ + [~ _1 __
48
282
282
~~] e2s
28
so that
y = ~e2t _ ~ _ ~t + [~e2(t2)  ~] OU(t  2).
4
4 2
2
2
9. The Laplace transform of the differential equation yields
.s:'{ }
1
y = (8+2)2+1e
271's
so that
y = e2(t271') sin tOU(t  21T).
10. The Laplace transform of the differential equation yields
.s:'{y} =
(8:
1)2es
so that
y = (t  l)e(tl)OU(t
 1).
2
3
8 +2
3
(71'S
+
+ 1
e
+e 371'S)
3 (8 + 2)2+ 32 (8 + 2)2+ 32 3 (8 + 2)2+ 32
=
350
Exercses 7.5
so that
y
+ ~e2(t37l")sin3(t
 3?T)OU(t  3?T).
3
12. The Laplace transform of the differential equation yields
1
e2s
+ e4s
s 1
'5 (s 
1 1
1)2 + 25 s  6
[1
+ '5 s 
1 1 ] (2S
1 + '5 s _ 6 e
so that
y
_~et
25
_ ~tet + ~e6t
5
+ [_~et4
25
+ [_~et2
+ ~e6(t2)] OU(t  2)
.z:{}
1 2 "(O)
1 3! /1'(0)
1 Po 3!
y ='2s3Y
+6"s4Y
+6"Els4e
so that
Ls/2
1 /1 ) 2 1 /1,( 3 1 Po (
y = '2Y (O x + 6"Y O)x + 6" El X Using y/l(L) = O and ylll(L) = Owe obtain
y = ~ PoL x2 _ ~ Po x3 + ~ Po (x _
4 El
6 El
6 El
2
Po (!:"x2 _ ~x3)
O < x < !:..
El 4
6
' 2
 { POL2 (~x _ !:___)
4El 2
12'
= Oand
y'(L)
= Owe obtain
y = ~ PoL x2 _ 2._ Po x3 + ~ Po (x _
16 El
12 El
6 El
2
O<
x
L
<x<L.
2 351
<2
+e
4s)
Exercises 7.5
15. The Laplace transform of the differential equation yields
1
~{Y}=82+w2
so that y(t) = sinwt. Note that y'(O) = 1, even though the initial condition was y'(O) = O.
Exercises 7.6
8~{X}
+ ~{y}
= ~{x}
8~{Y} 1 = 2~{x}
so that
1
Then
x
1
_et
3
1
_e2t
3
and
2
3
1
+ _e2t
Y = _et
8~{x}1=2~{y}+
81
8~{Y}  1 = 8~{x}
 2"
8
so that
~
{y}
83 + 782  8 + 1
8(8  1)(82  16)
1 1
8
1
173 1
53 1
16 ;  158  1 + % 8  4  160 8 + 4
and
1
8 t
y =   e
16 15
173 4t
+ e
96
53
e
160
4t
Then
x
1,
sy
st1 = st lIt
173 4t
15e + 192e
8~{X}
+ 1 = ~{x}  2~{y}
8~{Y}  2 = 5~{x}
352
~{y}
53
4t
+ 320e
Exercises 7.6
so that
8 
3 s2 + 9
::t{x} = s2 + 9 =  s2 + 9 and
5 .
3
x =  cos 3t   sm 3t.
Then
1
1
223
7 .
y = x  x = 2cos3t   sm3t.
4. Taking the Laplace transform of the system gives
1
s
+ (8 l)::t{y}
= 8 1
so that
5s  1
11
~+ 3 s 
3 (s 
1+
1
1)2
and
::t{x} _ 1  28
11 1 1
1
1
 3S(8  1)2  3 ~  3 s  1  3 (s  1)2 .
Then
X
1 t
= lIt  e  te
and
3
Y = 
lIt
4 t
+ e + te.
3
(s  3)::t{x} + (8  3) ::t{y} = 
so that
s  3
11
::t{x} = s(82)(83)=2~+2s2and
38  1
83
11
~ 2 s 
+ '3 s 
3'
Then
x
5 2t
3t
1 5 2t 8 3t
=  1 + e
 2e
and y =   e +e
2
(8 + 1)::t{x}  (8  l)::t{y} = 1
8::t{X} + (s+2)::t{y}
353
=1
Exercises 7.6
so that
:L{ } _
Y
8 + 1/2 _
8 + 1/2
82 + 8 + 1  (8 + 1/2)2 + (V3/2)2
and
:L{x} =
3/2
82 + 8 + 1
3/2
(8 + 1/2)2 + (V3/2)2'
Then
y =e
_/
t 2
V3
cos t
. V3
x = e t/2 Slnt
2 .
and
+ l):L{x}  :L{y}
(82
=2
+ (82 + 1) :L{y} =
:L{x}
so that
282  1
1 1
1
2
3.
2V2
+t  
Then
y =x
11
Slll V 2 t.
+ x = t1 +
2
3
r:
In sin v 2 t.
2v2
(8 + 1) :L{x} + :L{y} = 1
4:L {x}  (8 + 1) :L {y} = 1
so that
and
8
+3
8+1
2
(8 + 1)2 + 22 + 2 (8 + 1)2 + 22 .
:L {y} = 82 + 28 + 5
Then
x
= et cos 2t + 2et sin 2t and y = _et cos 2t + 2et sin 2t.
d2x
dt2
"2t2+ 2t
d2y
1
= _t2  2t.
dt2
2
354
Exercises
Taking the Laplace transform of the system gives
1
:t'{ x}
= 8~
and
1 4'
1 3!
+ 24 s5 + '3 s4
1 4!
1 3!
3 s4
:t'{y}=24 s5
so that
(S + 2) :t'{x}  2s3:t'{y} = O
so that
4
414
:t'{x} = (s _ 2)(s2 + 1) =
and
2s + 4
:t'{y}
5s
2
= s3(s2)(s2+1)
=;
81
5 s2 + 1  5 s2 + 1
2
s22s3+5s25s2+1
Then
4 2t
x = e
5
  cos t   sin t
5
5
and
y = 1  2t  2t
,
+ 51 e2t  56 cos t + 58 sin
i.
2s + 1
1
1
1 2
1
:t'{x} =  s3(s + 1) = ; + s2 + '2 s3  S + 1 '
Then
X
= 1+ t +
= '3te
t  e
t
and
1
t
'3x
11
= '3te
355
t
+ '3e
+5s2+1'
t
1
3
7.6
Exercises 7.6
12. Taking the Laplace transform of the system gives
2es
1 es
= 2 +8
+ (8+ l)..:t'{y}
so that
(J){ x } =
1/2
(81)(82)
1
= [281282
+e s
1
1]
1
(81)(82)
+e
s
[1
81
and
..:t'
es
8/41
=451282+e
8/2+2
s
s
1]
+ 82
(5  1)(8 2)
[l 3 1
~281
1]
+52
Then
x = ~et  ~e2t
+ [_etl + e2(tll]
y = ~et _ ~e2t
4
2
+ [1 
VU (t  1)
and
~etl
+ e2(tll] au (t  1).
+ 2(X2 
X'{
3XI
X~
2(X2  Xl)
Xl)
XI(O)=O
X~
(O) = 1
X2(0) = 1
x;(O)
= O.
+ (82 + 2) ..:t'{X2}
=8
so that
..:t'
+ ~ _1 __
5 82+ 1
and
356
~ _8_ + _4_ V6
5 82+ 6 5V6 52+ 6
Exercses
:L x
{2}
83 + 58 + 2 = ~ _8_
+ ~ _1_ + ~ _8
(82 + 1)(82 + 6)
5 82 + 1 5 82 + 1 5 82 + 6
7.6
2_ V6
5V6 82 + 6 .
Then
Xl
2
5
+
1
2
sin t   cos V6 t
5
5
4
5
2
5
=  cos t
and
1
5
Xl
4
re sin v'6 t
5y 6
2
;; sin v'6 t.
5y6
= k2 = k3 = 1, mI = m2
= 1,
 :L{X2} =1
:L{xl}  (2 + 82):L{X2} = 1
so that
Then
Xl
=
1.
r:
v'3
sin y 3 t
E(t)
= Ril
= Ril
and
= iz + i3.
X2 =
1.
v'3
sin h t.
= E(t)
+L2i~ or Lli2+Ri2+Ri3
10,000
+ 500:L{i3} = _
8,000
357
and L2i~+Ri2+Ri3
= E(t).
Exercises 7.6
so that
:t: .
8,000
80 1 80
{Z3} = s2 + 900s = 9" ~  9" s
+ 900
Then
.
Z3
(e) il
= i2 + i3
80 80 900t.
9"
 9"e
and
Z2
./
= 20  0.0025z3 
Z3
100 100
9
ge
900t
= 20  20e900t
i; + i~ + 10i2 = 120 
120OU(t  2)
+ 5i~ + 5i3 = O
10i;
gives
(s
. + S:t:{Z3}
. = 120 ( 1 + 10) :t:{Z2}
e _)28
+ 5(s + 1) :t:{i3}
10s:t:{i2}
=O
so that
(LJ
ol.
{z?J 
120(s
(3s2
+ 1)
+ lls + 10)s
( 1  e 2s)  [48
  60
 s + 5/3
s+2
+ 12]
s
( 1  e 28)
and
:t:{i}=
3
3s2
240
+ lIs + 10
=[
(1_e2S)
240
+ 5/3
_ 240] (1_e2S)
s +2
Then
iz
i3
 60e2(t2)]
OU(t 2)
and
(b) il
= i2 + i3 = 12 + 288eSt/3
 240e2(t2)]OU(t
 300e2(t2)]
6:t:{i2}
+ (s + 6):t:{i3}
358
= s2
50
+1
= _250
 2).
+1
OU(t  2)
Exercises
so that
::t i
{2}
50s
(s+2)(s+15)(s2+1)
= _ 20 _1_
13s+2
375 _1_
1469s+15
145 _s_
113s2+1
+~
_1_
113s2+1'
Then
and
i2 =  20 e2t
13
375 e1St
1469
+ 145 cos t + ~
113
113
,
25.
1,/
11 ,
30 2t
23 = 3sin t  '622 622 = 13e
250
+ 1469e
sin t
280
 113 cos t
1St
810 ,
113 sin t,
0,5i~ + 50i2 = 60
0,005i~ + i2  il = O
gives
,
,120
+ (s + 200) ::t{i2}
=O
so that
::t{i } _
24,000
2  s(s2
s+ 100
cos 100t 
~e100t
= 6 1
100
+ 200s + 20,000)
Then
i2 = ~ 555
and
~el00t
, =e 0005
,/ + 22
' =S
6
21
u.
22
sin 100t
+ i2 
il = O
gives
,
,60
2S::t{2} + 50::t{22} = 
200::t{iI}
+ (s + 200) ::t{i2}
=O
so that
::t{' }
6,000
22 = s(s2
61
+ 2008 + 5,000)
6
s + 100
 S;  S (s + 100)2 
(50)2)2
359
6)2
 5 (s
50)2
+ 100)2 
(50)2)2
'
7.6
Exercises 7.6
Then
6 6
6J2.
i2 =   _elOOt cosh 50J2 t  _elOOt
5 5
5
sinh 50J2 t
and
= i2+i3.
= dqjdt.
Using
Kirchoff's second law and summing the voltage drops across the shorter loop gives
1
(1)
so that
.
1
Il = E(t)
 q.
RlC
Rl
Then
.
dq
=
Il dt
and
l.
I3 = E(t)
 q  I3
RC
Rl
dq 1
Rl dt + Cq + Ri3
= E(t).
di3
dt
ilRl + L + R2i3
or
di3
L + R2I3
(b) Using L
= iR
+ Cq
Cq = O.
+~
{i3} =
5Ce1
__
8+1
es
(8 + 1)~{i3}  ~ {q} = 0,
so that
and
q(t) = 50e1e(tl)
360
Exercises
7.6
4e~ + e~ + 8el = O
e~ + e~ + 2e2 = O
gives
4 (s2 + 2).:t{el}
+ s2.:t{e2}
s2 .:t {el}
= 3s
=o
so that
or
Then
123
e2 =  eos ""t   eos 2t
2
vs
and
e~ = e~  2e2
so that
el =
(b)
1
4
2
y3
3
4
91
92
1
2
Mass
m2
m2
Mass
m1
position at about t = 0.66. The motion of the pendulums is not periodie sinee eos 2t/
period V3 1[", eos 2t has period
number.
1[",
361
J3 , which
J3 has
is not a rational
Exercises 7.6
(d1/
t= 1
:s t :s 30.
t= 2
t= 4
92
81
82
1
2
3'
4
5
6
7
8
9
10
0.2111
0.6585
0.4830
0.1325
0.4111
0.8327
0.0458
0.9639
0.3534
0.4370
0.8263
0.6438
l. 9145
0.1715
1.6951
0.8662
0.3186
0.9452
l. 2741
0.3502
t= 7
t=
(e) Using a CAS to solve el(t) = e2(t) we see that el = e2 (so that the double
pendulum is straight out) when t is about 0.75 seconds.
362
10
t=0.75
Xl(t) = 16sinBI(t)
and
Yl(t) = 16cosBl(t)
and
and
.:t'{f(t)}
= rltestdt+
(X)(2_t)estdt=
Jo
Jl
22es
8
tI/2.
7 . .:t'{e7t}
128
= _1_
8+7
8. .:t'{te7t}
9. .:t'{sin2t} =
(8:
7)2
i+4
8
10. .:t'{e3tsin2t}
= (8+3~2+4
2]
.
11. .:t'{t sin
2t} =  d [2ds 8 + 4
= (2 48 )2
8
+4
2
12. .:t'{sin2tOU(t  7T)} = .:t'{sin2(t  7T)OU(t 7T)} = 2e1fS
8 +4
363
17.
:::t1{
18. :::t1
19.
20.
s
} =:::t1
s2  lOs + 29
12e5S}
:::t1{
(s 
s5
5)2
+ 22
+~
(s 
2
5)2
+ 22
=tcos2t+~tsin2t
= (t  5)OU(t  5)
+ 1f 2 e S}
(1)1 { S
oL
(1)1 {
2
s+1f
L2s2 + n21f2
2e
s
1f
+ s+7i
2
2 e
S}
=oL
2
s+1f
= COS 1f (t
} _ ~ !:_:::t1{
 L2 n1f
21.
22.
:::t{teStf(t)}
23.
24.
:::t{l
n1f/L
}
s2 + (n21f2)/ L2
_lsin~t
Lnat
=  :sF(SS).
= eks
= ~:::t{eatf(t)}
eaTf(T)dT}
F(ssa),
= ek(sa)
F( s  a)
whereas
f(t)OU(tto)
26.
27.
28.
29.
2_  2_es _ ~e4s
s2
s2
364
Chapter
:L{etf(t)}
30. f(t)
e(sl) __ 1_e4(s1)
sl
(sl)2
= sint"i1(t  7T)  sintOU(t  3T) =  sin(t  7T)"i1(t 7T) + sin(t  37T)"i1(t 37T)
:L{f(t)}
= 2
:L{f(t)}
:L{etf(t)}
+ _1_e37rs
s2 + 1
l_e7rs
S2 + 1
= __
:L{et f(t)} = _
31. f(t)
(s1)2
7 Review Exercises
e7r(sl) +
(s1)2+1
=
e37r(s1)
(s1)2+1
= 2 + (t 
2)au(t  2)
1
+ 2e2s
s
_2_
sl
e2(s1)
(s1)2
1)
+ (t  2)"i1(t 
<P{f(t )} = 1  e
2 s +e1 2s
s2 s2
s2
eL
:L{et f(t)} =
1
.
(s1)2
e(sl) +
(s1)2
e2(s1)
(s1)2
:L{y}
(s _ 1)2(s2  8s + 20)
6
s4
6
169
_et
1
+ tet
13
6
169
 _e4t
365
cos 2t
+ _t5.
338
sin 2t.
2)
~ {}
s3 + 6s2 + 1
1
2s
2
2s
+ l)(s + 5)  s2(s + l)(s + 5) e  s(s + 1)(8 + 5) e
y = 82(S
6
13
_ (_~ . ~ + ~ . ]_ + ~ . _1
1_ . _1_) e2s
25 s 5 s2 4 s + 1 100 s + 5
+ _]_.
e2s
5 5 2 s + 1 10 s + 5
_ (~.~_ ~._1_
_1_)
so that
y = _~ + ~t2
25 5
+ ~et _
13e5t
50
100
~{y} =
s3 + 2 _ 2 + 2s + s2 es
s3(s  5)
s3(s  5)
=
2 1
2 1
1 2
125 ; 25 s2  5" s3
127
+ 125 s 
5
[37
1 12 1
1 2
 125 ; 25 s2  5" s3
37
l]_s
+ 125 s _
5 e
so that
y
127
=  2  t2  t1 2 + e
125
25
125
5t
 37)2 (t
125 25
1  1)2 + e
37
 1)  (t
5
125
1 2
~{f}
so that f(t)
= 6t.
s~{x}
4~{x}
+~{Y}
1
s
2" + 1
+ s~{Y} = 2
366
1
s
= 6 2"
5(t1J]
OU(t 1).
.:t'{x}
S2  2s + 1
s(s _ 2)(s + 2)
11
= 4 ~ + 8 s 
+ 8 s + 2'
Then
x
119
=  + _e2t + _e2t
4
and
Y = x
9 2t
1 2t
+ t = e
 e + t.
s2 .:t'{x}
+ s2 .:t'{y} = s _ 2
2s.:t'{x}
+ s2.:t'{y}
1
= s _ 2
so that
2
11
1
.:t'{x} = s(s2)2=2~2s2+(s2)2
and
.:t'
s  2 __ ~ ~ _ ~ 2_ + ~ _1_ _
4 s
2 s2
4 s  2
(s  2)2 .
Then
x
1
2
1
2
=   t
4
2
3 2t
2t
+ e
 te .
4
+ 2t.
.:t'{t}=
( 4
s+2
= s2(5s+2)
2)
s3
+ s2 10s+2
45
=~+
s2
+ 5s+1
=~+
Thus
i(t) = 9 + 2t
+ get/5.
dq
2 dt2 + 10 dt + 100q =
10  10OU(t  5).
.:t'{ }
q == 2 (s2
20
+ 20s + 200)
_ [2_ ~ _ 2_

10 s
10 (s
(1
 e
5S)
s + 10 _ 2_
10
] (1 _
+ 10)2 + 102 10 (s + 10)2 + 102
367
s2+ s+1/5'
e5S)
q(t) =  
 [~
10
 ~e10(t5)
10
cos 10(t  5) 
~e10(t5)
10
43. Taking the Laplace transform of the given differential equation we obtain
.:t'{y} = 2wo
ElL
(.!:_.
4885
4! __ 1_ . ~ + _1_ . ~ eSL/2)
12086
12086
+ Cl . ~ + C2. 3!
283
684
so that
2wo [L 4
1 5
1 (
y = El L 48 x  120x + 120 x
where y"(O) = Cl and y"'(O)
ci 2
C2 3]
"2L) 5 OU( x "2L) +"2
x +"6x
= O and
. Cl = woL2/24El,
C2= woL/4El.
Hence
.:t'{ } = Cl . ~
y
284+4
+ wo (~_
+ C2. _4_
484+4
8El8
8 1
(81)2+1
.
 cosxsmhx)
8+ 1
(8+1)2+1
wo
+ 4E/1
)"
.
cosxcoshx)
wo
 8E/1
wo
 cos xcoshx).
(b) In this case the boundary conditions are y(O) = y'(O) = O and Y(7r) = y'(7r) = O. If we let
Cl = y"(O) and C2= y"'(O) then
368
:t{ } =
y
Cl . ~
2 54
+ C2. _4_ +
4
Y(7T)
[Sin
= O and
54
+4
Wo . _4_ esrr/2.
4El
54
cos
. 11X )
SlD
(x  ~) cosh (x  ~)  cos (x  ~)
Y'(7T)
+4
we get
cl.
. h
+ 4"
C2( .
= "2
SlDX SlD
X
SlDX COSh X
+ 4V;l
Using
+4
sinh
(x  ~) ] q (x  ~)
= O we find
Wo sinh
El sinh
11"
Cl=
Wo cosh
El sinh
11"
C2=
Hence
wo
sinh!I.
y = 2El ~h
sin
2 sinxsinhx
11"
+ 4V;l [sin
Wo
cosh!I.
El ~h
4
sin
2 (sinxcoshx
11"
 cosxsinhx)
369
____
Equations
Exercises 8.1
x' =
2. Let X
(! 5)
8
= (~). Then
x' = (:
3. Let X ~
G)
4. Let X ~
G)
7)
o
X.
Then
x' =
4 9)o
1
103
o
o
x' ~ ( :
1
x~ (~)
X.
Then
1
5 Lec
. X.
Dx
Then
1
x'~U Dx+
1
1
(t;t2)
370
U)
+ .:
Exercises 8. 1
6. Let X ~
7.
Then
dx
di = 4x + 2y + et;
dx
8. dt
9.
G)
= ~IX + 5y 
dx
di
=X
9z  8e t;
Y + 2z
+ e
 3t;
dz
= 2y + 3z + e5t  3e2t
dt
dy
t
dz
t
dt = 3x  4y + z + 2e '+ t; dt = 2x + 5y + 6z + 2e  t
dy
dt
= 4x + y + z + 2e5t;
dx
10. d =3x7y+4sint+(t4)e4t;
dy
=x+y+8sint+(2t+1)e4t
11. Since
X' = ( 5 ) e5t
10
and
we see that
X'=
12. Since
I
X =
5 cos t  5 sin t ) t
2 cos
t  4 sin t
and
we see that
X t :=
13. Since
( 3 4)
( 3 4)
4 7
4 7
X=
XI
10
5t
X.
t  5 sin t)
(  22 45) X = (52 cos
cos t  4 sin t
(22 45)
( 1 1/4) X = (3/2)
= (1 1/4)
1 1
et
X.
1 1
we see that
(5)
e3t/2
3
X,
14. Since
371
~) X.
et
+(
4)
4
te'
Exercises 8. 1
15. Since
we see that
16. Since
X'=
cos,
)
~ sin t  ~ cos t
 cost  sint
we see that
and
1)
(
COS,
)
O X = ~ sin t  ~ cos t
 cos t  sin t
O 1
1
X'~ ( : 1 ~) X
2 O 1
17. Yes, since W(XI, X2) = _2e8t =1 Oand X and X2 are linearly independent on
00
18. Yes, since W(XI, X2) = 8e2t =1 Oand X and X2 are linearly independent on
00
00
21. Since
we see that
22. Since
we see that
23. Since
372
Exercises
8. 1
we see that
24. Since
X' =
p
3 cos
O
3t)
and
3 sin 3t
1
4
2 3)
2 O X,
6
1 O
( 1)
4 sin 3t
3
( 3 cos 3t )
3sin3t
we see that
25. Let
Then
x; ~
X3~ m'~AX3'
and W(XI,X2,X3)
00
20
i O so that
26. Let
1 r ) e v'2t ,
Xl = (
1 v2
= (
1 +
j2
) ev'2t
373
Exercises 8. 1
and
Then
x:2 = ( 2}2
}2
00
) ev'2t
= AX 2,
Exercises 8.2
l. The system is
x'=
and det(A  >"1) = (>.. 5)(>..+ 1)
( : _~ I ~)
e ~)x
1 ~2 ~)
so that
K1 ~
e).
Then
X = ci
( 21)
2. The system is
x'=
e5t
(1)
+ C2
e.t
e ~)x
e ~I~)=C~I~)
so that
Kl~(~)
2
1
2 ':
==> (1
1
1
O
O o
lO)
Io
374
so that
K2
(11).
Exercises 8.2
Then
X =
elle( 2)
+ e2
(1)1 e .
4t
3. The system is
X'  (~~2 ~) X
and det(A  Al) = (A  l)(A
+ 3) =
O. For Al = 1 we obtain
so that
Kl
so that
K2 = (21) .
(25)'
3 we obtain
1 2 1 O) :=::;. (1 2 1 O)
( 5/2
5 1O
O O 1O
Then
X =
el ( 52) et + e2 (2)1 e 3t .
4. The system is
X' =
and det(A  Al)
=
(
For A2
(A
(5/2
3/4
+ 1)(2A + 7) = O. For Al
2)
2
= 7/2
1
210):=::;.(1210)
3/4 3/2 1 O
O O 1O
we obtain
Kl=(21)'
sothat
= 1 we obtain
( 3/23/4
K2 = (43)'
so that
1
Then
X
ci (  2)1 e 7t/2 + e2 ( 4)
3 e.
5. The system is
X' = (10
5) X
12
For A2 =
10
we obtain
1/4 lO)
( 208 52 1 OO):=::;. (1O O O
1
375
so that
Kl = (52)'
Exercises 8.2
Then
6. The system is
X' = (6 2) X
3
210)
( 6
3
1 1 O
(1 1/310)
==>
so
that
Kl =
(31).
For A2 = 5 we obtain
( 31 6210)
I O
==>
(1 210)
O
so that
K2 =
(21),
Then
7. The system is
XI=(~ ~ ~)X
O 1 1
and det(A  Al)
. ~
so that
m, m, m'
m m m,'
x ~ '1
KF
andKF
e"+~
"+02
8. The system is
X' = (~ ~~ :) X
052
and det(A  Al) = (2  A)(A  5)(A  7)
= O. For Al
376
= 2, A2 = 5, and A3 = 7 we obtain
Exercses 8.2
so that
= 1, A2
= 3, and A3
= 2 we obtain
so that
so that
= O. For Al
= 1, A2
= 1/2,
and A3 = 3/2
we obtain
Kl~(_D, K2~C1D,
so that
X ~ c
(D
' + '2
+ 5)(6
en
and
,'/2 + '3
 A) = O. For Al
377
K3~(_D,
(D
,31/2
Exercises 8.2
so that
+ 1/2)(A
 1/2)
K1=
= O. For Al
(~)
= 1/2
K2 =
and
and A2
= 1/2 we obtain
e) ,
so that
If
X(O)
then
C1
== 2 and
C2
= (:)
= 3.
cn'
m cn,'
H), m,
KI ~
K, ~
so that
x ~CI (
~)
,"
+02
X(O) ~
C1
15. X = ci
= 1,
C2
= 5/2, and
(~:~:~~:~)
e8.58979t
C3 =
C1
~~4~;~~:3)
e2.25684t
+ C3
(=~:~~~~~~)
eO.046632lt
0.359815
0.615458
0.35995
0.0312209
0.280232
0.262219
0.949058
16. X
and K3 ~
1/2.
+ C2 (
0.239535
0.836611
e5.05452t
= 1 we obtain
a" + C3
If
then
= 3, and A3
= 2, A2
+ C2
0.275304
0.162664
+ C3
0.195825
0.176045
0.826218
0.346439
0.0508861
0.338775
0.31957
0956lt
e2.92362t
continued
378
Exercises 8.2
0,313235
0,301294
0,64181
0.466599
0,31754
+C4
+ C5
e2,02882t
e o, l55338t
0,222136
0,173787
0,0534311
0,599108
0,799567
17. (a)
(b) Letting
Cl
1 and
C2
= O we get x = 5e8t, y
y = tx, x > O. When Cl = 1 and C2 = Owe find y = tx, x < O. Letting Cl = Oand C2
get x = elOt, y = 4elOt.
Eliminating the pararneter we find y = 4x, x > O. Letting
and
C2 =
1 we find y = 4x, x
< O.
(e) The eigenvectors K = (5,2) and K2 = (1,4) are shown in the figure in part (a).
and
C2
= 1 and
C2
we find y
x
Cl
= x, x > O. When
Cl
= Q and
C2
= 1 we find y = z ,
< O.
= O we get x =  2e7tj2,
y = e7tj2. Elimnatng the parameter we find y = ~x, x < O.
When Cl = 1 and C2 = O we find y = ~x, x > O. Letting
ci = O and C2 = 1 we get x = 4et, y = 3et. Eliminating the
parameter we find y = ~x, x O, When Cl = Oand C2 = 1 we
find y = ~x, x < O.
In Problem
4,
letting
Cl
1 and
C2
379
= 1 we
Cl
=O
Exercises 8.2
19. We have det(A  ,Al)
A solution of (A  ,A1I)P
= K is
P = (~)
so that
= (,A + 1)2
= O. For ,Al = 1
=K
is
P=
so that
X = ci
21. We have det(A  ,Al)
we obtain
C).
K=
A solution of (A  ,A11)P
(~)J.
+C2 [C)t+
X=q (~)
C)
:'
+ C2
C~5)
[C) u:' + C~5)
et]
A solution of (A  ,A11)P
=K
is
C).
P (1~3)
=
so that
X C)
= C1
= (,A  6)2
e2t
+ C2
= O. For ,Al
[C)
=K
+(
= 6 we obtain
K=
A solution of (A  ,A11)P
te2t
(~) .
is
so that
380
1~3)
e2t]
Exercises 8.2
23. We have det(A  Al) = (1  A)(A  2)2 = O. For Al = 1 we obtain
K~(J
For A2 = 2 we obtain
K, ~
and K3 ~
(J
Then
+ 1)2
= O. For Al = 8 we obtain
K~m
For A2
1 we obtain
K, ~ ( :)
and
K3 ~
H)
Then
K~(=n
For A2
= 5 we obtain
K~CD
A solution of (A  All)P
= K is
5/2)
P = 1/;
381
Exercises 8.2
so that
KI~m
For A2
=2
we obtain
K~(D
A solution of (A  A2I)P = K is
(n
p~
so that
X ~ '1
el +02 ( 
D ,"
 1)3 = O. For Al
+03 [(  ~ )
te"
+(
n
e"].
= 1 we obtain
K~(D
Solutions of (A  A11)P = K and (A  A11)Q = Pare
p~m Q~n)
m (D m [(D m n) "~l
and
so that
X ~ '1
+02 [
te' +
e'] +O,
K~m
382
t;"
te' +
Exercises 8.2
Solutions of (A  )l}l)P = K and (A  )l}l)Q
= Pare
p~m Q~m
and
so that
X~
'1
'+'2
+~
t'+
C).
P=
C)
A solution of (A  All)P = K is
so that
X
= el (~) e4t + e2 [
G)
te"
C)
e4t]
If
then el
= 7
and e2 = 13.
+ l)(A 
KI~cn
For A2 = 1 we obtain
KF(D andKF(D
so that
x ~ '1 ..
el
+ '2
If
X(O)~
m m
m
383
e' + '3
e'.
Exercises 8.2
then
el
= 2,
e2
= 3, and
e3 = 2.
Kl=
K2=
and
K3=
> O. When
el =
1 and
e2 =
we find y = x, x < O.
Letting c = 1 and
When c
= 1 and
e2 = Owe
e2 = Owe
find y
= x,
> O.
x < O.
In Problems 3346 the form of the answer will vary according to the choice of eigenvector. For exarnple,
in Problem 33, if Kl is chosen to be (
cos t
) e4t + e? (
sin t
) e4t .
X = el (
2 cos t + sin t
 2 sin t  cos t
= A2 
8A
+ 17 = O. For Al = 4 + i we obtain
384
Exercises
i)
KI= ( 2 +
5 '
so that
x, = (2+5
i)e(4+i)t=
e4t+i(cost+2sint)
(2Costsint)
5 cos t
e4t,
5 sin t
Then
X = el (
2 cos t  sin t ) 4t
( 2 sin t + cos t)
4t
e + e2
e ,
5 cos t
5 sin t
( li)
2
so that
XI=
1  i ) it
e = ( sin t  eos t ). +z.' ( cos t  sin t ) ,
2
2 cos t
2 sin t
Then
X
35. We have det(A  Al) = A2
= el (
sin t  cos t )
(  cos t  sin t )
+ e2
'
2 cos t
2 sin t
8A + 17 = O, For Al = 4 + i we obtain
K =
( li)
2
so that
Xl = ( 1  i ) e (4+ ~')t = ( sin t  cos t ).. e4t + z' (  sin t  cos t ) e4t,
2
2 cos t
2 sin t
Then
X = el (
36. We have det(A  Al) = A2
sin t  cos t) 4t
(  sin t  cos t ) 4t
e + e2
e ,
2 cos t
2 sin t
so that
COs3t+3sin3t)
2 cos 3t
5t
e + e2
385
(Sin3t3cos3t)
2 cos 3t
5t
e ,
8.2
Exercises 8.2
37. We have det(A  Al)
= A2 + 9 = O. For Al = 3i we obtain
Kl= ( 4 + 3i)
5
so that
Xl =
( 4 +5 3i)
Then
X = el
38. We have det(A  Al)
3it
5 cos 3t
+ e2
5 cos 3t
5 sin 3t
(4 sin 3t + 3 cos 3t ) .
5 sin 3t
= A2 + 2A + 5 = O. For Al = 1 + 2i we obtain
Kl =
e~
2i)
so that
=(2cos2t2sin2tcos2t)et+i
Then
X
el (
2 cos 2t  2 sin 2t ) t
e
cos 2t
For A2
=i
+ 1) = O.
2 cos 2t + 2 sin 2t )
.
et.
sm2t
For Al = O we obtain
we obtain
so that
Then
X = el
1)
O
O
+ e2
(Sin
t)
 sin t
cost
386
+ e3
(  cos t )
c~s t .
smt
t
Exercises
40. We have det(A  Al)
For A2
= (A + 3)(A2
2A
+ 5) = O. For
Al
= 3 we obtain
1 + 2i we obtain
so that
X2=
Then
X=C1
2 cos 2t
O)
2
For A2
 2 cos 2t + sin 2t )
3sin2t
et+i
2 sin 2t
(  2 cos 2t + sin 2t )
e3t+c2
3sin2t
et+c3
(  cos 2t  2 sin 2t )
3cos2t
e',
2 cos 2t
= (1  A)(A2
2A
2 sin 2t
+ 2) = O. For
Al
= 1 we obtain
= 1 + i we obtain
so that
Then
X = C1
O)
(
2 et +
(1
C2
8A
cos t )
s~nt et
 SIn t
+ 20) =
+ C3
( sin t )
COS t
et.
cost
O. For Al = 6 we obtain
387
8.2
Exercises 8.2
For '\2 = 4 + 2i we obtain
en
K2~
so that
X2 = (~)
e(4+2i)t
= ( Si~2t )
(C~'2t)
t+i
2 cos 2t
e4t.
2 sm 2t
Then
X=Cl
O)1
(O
For '\2
2
+ 3i
e6t+c2
sin 2t )
O
2 cos 2t
e4t+c3
(  cos 2t )
_o
2 sin 2t
e4t.
'\1 = 2 we obtain
we obtain
4 + 3i)
K2 =
5
O
so that
X2=
4 5+ 3i)
.
e(2+3t)t=
(4 cos5cos3t
3t  3 sin 3t)
(4 sin5sin3t
3t + 3 cos 3t)
e2t+i
e2t.
Then
X
=q
28 )
5 e 2t + C2
25
For '\2
COS
(
e 2t
+ C3
sm
O
= 2i we obtain
388
e.2t
Exercises
so that
X2 =
22i)
1
e2tt =
(2Cos2t+2sin2t)
cos 2t
cos 2t
+i
(2Cos2t2Si112t)
sin 2t
sin 2t
Then
O)
X
C1
(  2 cos 2t + 2 sin 2t )
cos 2t
Por "\2
e2t + C2
1
+ C3
(  2 cos 2t  2 sin 2t )
sin 2t
.
~~
~~
= 1 we obtain
= 5i we obtain
1 + 5i)
1
(
1
K2=
so that
X2 =
1 + 5i )
1
tt
( cos 5t  5 sin 5t )
cos 5t
+i
+ C3
( sin 5t + 5 cos 5t )
sin 5t
.
sin 5t
cos 5t
Then
X = C1
25 )
7 et
+ C2
( cos 5t  5 sin 5t )
cos 5t
cos 5t
If
U)
X(O)~
then
C1
C2
= 1 and
c3
= 6.
,,\2 
10,,\+ 29 = O. Por
K1 =
so that
Xl = ( 1 ) e (5+2i)t =
1  2i
"\1 =
5 + 2i we obtain
e~
2i)
cos 2t
) e5t
cos 2t + 2 sin 2t
+ z. (
sin 2t
) e5t .
sin 2t  2 cos 2t
and
.
cos 2t
)
X = C1 (
cos 2t + 2 sin 2t
5t
+ C3 (
389
sin 2t
)
sin 2t  2 cos 2t
5t
8.2
Exercises 8.2
If
X(O) = (:)
then
CI =
2 and
C2 = 5.
47.
Xl
YI = Y2
Y2
10YI
+ 4Y3
Y3 = Y4
y~ = 4YI  4Y3
or
o
y'=
10
O
4 O
4
Y.
Using a CAS, we find eigenvalues v'2i and 2y'3i with corresponding eigenvectors
:r=v'2i/4
1/2
:r=v'2i/2
1
1/2
O
390
:r=v'2/4
i
:r=v'2/2
O
Exercises
and
V3i/3
2
::::V3i/6
2
V3/3
O
V3/6
8.2
Thus
v'2/4
O
v'2/2
O
i ' and
C4
YI
CI
g,
C2
= O,
= O. Thus,
(t) = 
y;
sin ht
X2(t) = Y3(t) = 
V;
sin ht
 ~
Xl
(t) =
YI
V;
sin 2v'3t
sin2v'3t.
X~ = 10XI + 4X2
X~ = 4XI  4X2
or
X" = [10
We assume solutions of the form X
and 12,
WI
= .)(
2)
= v'2 and
4] X.
4 4
= V coswt and X
W2
)(12)
391
= 2V3.
Exercises 8.2
are
and
Then, the general solution of the system is
X = ci
cos ht
[~]
+ C2[~]
+ C3[ ~]
sin ht
cos 2v'3t
+ C4[ ~]
sin 2v'3t.
X(O) = [~]
imply
ci =
V2 sin ht + /3
10
V2.
Thus
sin 2v'3t
)3.
X2(t) = smht
 sm2/3t.
10
49. (a) From det(A  Al) = A(A  2) = O we get Al = O and A2 = 2. For Al = Owe obtain
1 1
1110
2 we obtain
(
For A2 =
I O)
=:}
(1 1
0010
I O)
so that
Then
X=ct
so that
( 1) (1)
1
+C2
1 e.2t
x =
Cl
+ C2e2t, Y = Cl + C2e2t or
y =x
392
K 1 = (11)
K2~ (:)
Exercises
so that
x = C! (  ~ ) + C2 [ (
 ~ )
= x.
and
393
t + (  ~) ] .
8.2
Exercises 8.2
The generalsolution of the system is
C2te2t
+ e e2t
C2e2t
x=
C3e2t
C5te2t
+ C4e2t
C5e2t
=e
e2t
+ C2
O
1
te2t
+ C3
e2t
O
1
+ C4
e2t
+ C5
e2t
te2t
e2t
O
1
= eKle2t
+ C2
Klte2t
e2t
O
O
O
O
+ C3K2e2t + C4K3e2t + C5
K3te2t
e2t
O
1
There are three solutions of the form X
the form X
= Kte2t + Pe2t
Ke2t,
See (12) in the texto From (13) and (14) in the text
(A  2I)K1 = O
and
394
Exercises
This implies
O O O
PI
O O O O O
O O O O O
P2
O O O O 1
O O O O O
P4
P5
O 1
so
P2 =
1 and
P5
= O, while
P3
Choosing
PI
= P3 = P4 = O we
1
O
p=
O
O
O
Therefore
a solution
is
x=
Repeating
O te2t
e2t.
for K3 we find
O
O
p=
O
O
1
so another
solution
is
x=
O te2t
+ 2y)2 = 4 sin2 2t
+ 2y
e2t.
=4
4y2
and
or
395
X2
+ 4xy + 8y2
= 4.
have
8.2
Exercises 8.2
This is arotated conic section and, from the discriminant b2
4ac
curve is an ellipse.
=O
Exercises 8.3
1. From
X'
( 3 3) X+ (4)
2 2
1
we obtain
Then
and
so that
U=
q,lF dt
and
Xp
= <llU
q,l
2
= ( et
J ( 11) = (l1t)
= ( 11) t + (15) .
11 10
=
5et
and
q,l
dt
5et
2. From
we obtain
Then
so that
U
q, I F dt
(2tet)
2tet
and
396
3 t
e
2
_let
dt = (2tet
2tet
+ 2et)

2et
Exercses 8.3
3. From
X' (3 5) X + ( 1)
=
3/4 1
et/2
1
we obtain
Then
so that
U
<p1F dt
J ( 3 t)
~e1i
dt
and
Xp = <pU =
4.
13/2)
( 13/4
te
t/2
+(
(3~4~tt)
(15/2)
/
9 4
t/2
e.
From
X'
= (2
1)
2
sin 2t )
2 cos2t
we obtain
sin 2t )
Xc = el (
2 cos 2t
+ e2 (COS
2t ) 2t
e .
2 sin 2t
2t
Then
<p
_e2t
(
sin 2t
2e2t cos2t
cos 2t)
2e2t sin 2t
e2t
and
so that
U=
<p1Fdt =
and
Xp = <pU =
J (~
1
<p1
_le2t
~e2t
cos 4t ) dt =
"2 sin 4t
e2t
cos 2t
1
e2t sin 2t
4
(i
sin 4t )
1
;o; cos 4t
5. From
X'
sin 2t
= (
O 2) X + ( 11)
1 3
we obtain
X, = elle
( 2)
(1)
+ e2 1
Then
and
1
<p.
397
et
e 2t .
t
e
2t
e
2t
e .
cos 2t)
Exercises
8.3
so that
and
6. From
X'=(_~
~)X+C~3t)
we obtain
x, = el
et
(~)
+ e2
e)
e2t.
Then
cp _ (2et t
e
e2t)
2t
e
and
cp1
et)
~2t
t
(
_:2t
so that
U=
cP 1 Fdt 
J(
2et  e4t
_2e2t + 2e5t ) dt
( 4)
2e
t
1 4t)
+ ".te
= (e 2t  se
2 5t
and
7. From
we obtain
X,
el
1 e3t + e2
(2)
1 e
Then
4e3t
cP
= ( e
3t
_2e3t)
3t
e
so that
3t.
and
cP
1
3t
6"e
_ e3t
3t
(2t e 3t  3"2e3t)
3t
)
dt
=
2t e 3t :3 e3t
6te
6te
and
Xp
= cpU =
O t + (4/3)
4/3 .
( 12)
8. From
398
Exercses 8.3
we obtain
X
= CI
( 4)
1
e3t
+ C2
Then
q,
= (
4e3t
e
3t
so that
and
9. From
we obtain
Then
so that
and
10. From
we obtain
Then
so that
399
(2)
1
e 3t .
Exercises 8.3
and
3).
xp = eu = ( 5
11. From
XI _
 (O
( sec t )
O
O X+
1 )
we obtain
cos t )
+ e2
= el ( sin t
Xc
(sin t )
_ cos t .
Then
<P=
cost
( sin t
sin t )
cost
<P
and
1
sin t )
 cost
(cos t
. t
Slll
so that
and
t cos t
Xp
12. From
( 1 1) X+ (3)
X' 
et
we obtain
X = el (
c
sin t ) t
e
cos t
t)
+ e2 ( cos
. te.
sin
Then
<P = (
 sin
') t
cos t
COS t
. t
sin
<P 1 = (  sin
an
cos t
t)
.
cos
sin t
e t
so that
U =
J <P
1F
dt
3 sin t
3 cos t
+ 3 cos t)
+ 3 sin t
and
Xp = <pU
13. From
I
XII
dt
( 3)
(3 cos t + 3 sin t)
3 sin t  3 cos t
e.t
(1 1) X + (c~St) e
sin t
we obtain
X=cr
c
sin t ) t
e+c2.
cos t
400
cos t )
te.
sin
Exercises 8.3
Then
<I>= (
sin
t)
cos
sin t
cos t
so that
u=
_]
<I>] F dt =
J (~)
and
cos
Xp = <I>U = ( .
smt
14. From
XI =
<I>
and
( 
cost) et
sin t
dt = (~)
t)
te .
(2 2) X+e(1) 1
8 6
sin
cos t
2t
we obtain
Then
1 2t+1)
<I>= ( 2
4t
+1
2t
_]
<I>
and
((4t+1)
21
2t
+ 2 In t )
_ In t
dt
2t+1)e2t
so that
U =
<I>] F dt =
J(
and
2t
X
p
= <I>U = (
4t
+ In t 
2t In t)
+ 3In t  4t In t
_ 2t
15. From
we obtain
Xc = CI ( cos t ) + C2 ( sin t )
 sin t
cos t
Then
cos t
<I>= (
 sin t
sin t )
t
cos t
so that
U =
and
Xp=<I>U=
<I> I F dt =
and
J (
tan2
tan t
t ) t+. (
cos
.
(  sin
t
<I>
t)
t)
sin
sin t tan t
401
= ( cos t  sin
sin t
dt =
t)
cos t
(t  tan t)
In I sec ti
Exercises 8.3
16. From
we obtain
cos t )
Xc = C1 (
.
sm t
Then
cp=
cos
(  sin t
U=
t)
sin
cos t
+ C2 ( sin t )
cost
1 (COS t
and
cp =
O)
dt =
 sin
t)
cost
sin t
so that
and
cp1F dt =
J(
ese t
In I ese t  cot ti
17. From
XI =
(11/21 2) X+ ( sect
ese t ) e
we obtain
2 sin
Xc = cr (
t)
cost
+ C2 ( 2 cos
.
smt
t)
e.
Then
cp= (
2 sin t
cos t
2 cos t) t
e
 sin t
Md
cP
1
= ( ~sin t
~ cos t
cos t) s
e
 sin t
so that
U =
cP 1 F dt =
J ~
(
J2
) dt = (
2
cos t  tan t
~ In I sin ti  In I sec ti
and
(COS1.
t ) In I sin ti + (  2 cos
. t ) In I sec r],
 '2 sin t
sm t
(1
we obtain
Xc = C1 (
Then
cp =
(COS t 
sin t
cos t
cos t  sin t )
cost
cos t + sin t )
sin t
+ C2
and
402
( cos t + sin t )
.
sin t
cP
= (
sin t
cost
t)
cos + sin
sin t  cost
Exercises
so that
1
dt =
2 sin t  cos t
and
Xp = cpU = (
+ tan ti)
19. From
we obtain
Then
1
cp =
e2t
1 e2t
so that
and
20. From
x' =
(~
:
1 1
=~) X + ( ~ )
1
2et
we obtain
Then
403
t
e2t
e
e2t
8.3
Exercises 8.3
so that
U
.
q>lF dt =
(tet +
2et
2) dt = (te
 et
2et
21t e 2t
 t e 2t
+ 2t)
+ ;re
1 2t
and
Xp=q>U=
1/2)
1
( 1/2
21. Frorn
,(3
X=
we obtain
q> (_e
=
1) X +
1
e2t)
2t
e
t
4t
(3/4)
1
3/4
t+
(2)2
O
et+
(2)2
2
teto
(4e2t)
4e4t
'
and
X = q>q>l(O)X(O)+ q>
r t q>lFds
Jo
= q>.
(O) + (eq>.
2t
2t
+ 2t + 2t 
1)
1 1
we obtain
q>=(11
l+t)
t
q>_l=(t
'
l+t)
1
1
'
and
X = q>q>l(l)X(l)
+ q>
q>IFds =
e (:)
+ e C~t)
= (~)
t  (~)
23. Solving
2 A
3
1 = A2  1 = (A  l)(A
+ 1) = O
1 2  A
we obtain eigenvalues Al = 1 and A2 = 1. Corresponding eigenvectors are
1
Thus
404
C)
lnt.
Exercises
8.3
Substituting
al
al =
1 and
X(t)
2b1 = 5,
h = 3. Then
= el (:
) et
+ e2 (;)
et
+ ( ~) .
24. Solving
6 A
3 A
+ 14
= A2  9A
Thus
X, = el
(A  2)(A  7)
=O
( 1)
4
e 2t
+ e2
(1)
1 e 7t .
Substituting
4al
+ 3bl
b2
+ 4 = O.
= 6. Substituting these two
give al = ~ and bl = ~. Then
e2t + e2 (~)
25. Solving
405
e7t + ( :) t + (
J) .
Exercises 8.3
we obtain the eigenvalues
)11
= 3 and
A2
Thus
1)
x c = c1 ( 3
+ c2
e3t
(1)
9 e 7t ,
Substituting
X, =
b
( al)
3a1
+ 3'b1
'9a1
+ 5b1 = 10
=3
1)
( 3 e3t
el
Then
+ C2 ( 91)
e7t
( 55/36 )
19/4
et,
26. Solving
1A
1
=A2+4=0
1 A
1
1 + 2i
and
K2
= (
5 ),
1  2i
Thus
5 cos 2t
)
Xc = Cl (
cos 2t  2 sin 2t
+ C2 (
5 sin 2t
)
2 cos 2t + sin 2t
Substituting
al
+ 5bl + a2 + 1 =
al
+ b: + oz = O
406
'
Exercises
27. Let I =
= 3,
oz
= 2/3,
al
= 1/3,
and
br = 1/3. Then
5cos2t
) + e2 (
5sin2t
) + (3)
= el (
cos 2t  2 sin 2t
2 cos 2t + sin 2t
2/3
C~)
cost
+ (1/3)
1/3
so that
and
Then
3e12t )
e12t '
U = IJ>  1Fdt
= (lOe2tSint)
30e12tsint
dt = (2e2t(2sintcost))
26ge12t(12sint  cost)
and
Ip = IJ> U =
so that
1= (1) ee1
fg.
332sin t 29
( 276sin t 29
2t
+ e2
76cos t )
29
168cos t
29
3) e + I .
( 1
12t
= A2
2A = A(A 2)
= O.
Kl = [~]
Thus
and
[ 1] [1]
X, = el 1 + e2
If we substitute
407
1 e 2t .
sint.
8.3
Exercises
8.3
[0] [1 1]
1
[al]
bl
[3]
5
br+3]
+ bl 
[al
al
5 '
which cannot be solved for al and bi. The difference in the systems is that in Problem 23 the
homogeneous system does not have a constant solution, whereas in this problem it does.
(b) Trying a solution of the form
1.
= a2
+ bl  5) + (a2 + b2)t = b:
+ bl
and bl = 4. Thus
29. (a) The eigenvalues are O, 1,3, and 4, with corresponding eigenvectors
and
(b)
e=
( ~6
4
2et
3e3t
et
e3t
2t
e3t
~e")
<pl
e
1
3
_,
1 4t
"3e
(e) <Pl(t)F(t) =
408
1 t
"3e
O
2
"3e
4t
"3
2et
2
"3e
3t
"3
2t
e
_\ e3t
1 4t
"3e
Exercises 8.3
4>(t)C + 4>(t)
J 4>l(t)F(t)dt
409
Exercises 8.4
Exercises 8.4
1. For A
= (~ ~) we have
~),
A3=AA2=(~
~)(~ ~)=(~
~),
A =AA3=C
~)(~ ~)=(~
16)'
= (~
20k)
for
= 1, 2, 3,
o o o
Thus
t2
t3
2!
3!
l+t+++o.o
(
and
(2t)2
(2t)3
2!
_At=(e
2. For A
l+t+++
0)
e2t
3!
= (~ ~) we have
410
~)
) (e
=
Exercises 8.4
and so on. In general
k
{A,
1,
k
k
= 1, 3, 5,
= 2, 4, 6,
o o o
o
o o o
Thus
e
At
A
A2
= 1+ t
+ t
1!
2!
= 1+ At
A3 3
+ t
+ooo
3!
+ 2.It2 + 2.At3 + o o o
2!
3!
= 1 (1 + 2.
t2 + 2.t4 + o o o) + A (t + 2.t3 + 2.t5 + o o o)
21
4!
3!
51
cosh t
sinh t )
= Icosht + Asinht = (
sinh t cosh t
and
eAt = (COSh(t)
sinh(t)
sinh( t))
cosh(t)
= (
cosh t
sinht
 sinh t)
cosht
3. For
A=( ~ ~ ~)
2
2
2
A2 = (~
2
2
we have
Thus, A 3 = A 4 = A 5
eAt = 1 + At =
o o o
~) (~
2
2
2
~ ~)
O O O
O and
O o1
O) + (
O
O O 1
A~U
O
O
1
') ('+1
tt
t
t
2t 2t 2t
4. For
we have
~) = (~
2
n
411
t
2t
t
t+1
2t 2t + 1
Exercises 8.4
Thus, A 4
= AS = A 6 = ... = O and
eAt
= 1 + At + 21A 2t2
e 0) e
= o
o o
3t
5t
o
o o
t o
0) +
o
o o o =
~t2 o o
0) ( 1
3t
~t2 + 5t
x= ( eot
G~)=
e~t )
ci (~)
+ C2
(~
) .
x = (COSht
sinh t
t)
sinh
cosh t
(q)
t)
= c (COSh
+ C2 (sinh
sinh t
C2
t) .
cosh t
X =
t +t
2t
t)
t +t 1
t
2t 2t + 1
(Cl)
C2
= C!
( t +t 1) +
2t
C3
3t
( ~t2 + 5t
9. To solve
412
C2
(t)t
+ 1 + C3 (t)
2t
2t
+1
Exercises 8.4
we identify to
= 0, F(s) = (_~), and use the results of Problem 1 and equation (5) in the texto
X(t) = eAtC+eAt
{t eASF(s)ds
Jto
) (e e2s0) ( 3) ds
{t
= (~
e2t Jo
{t ( 3es
e2t Jo
_e2s
1
ds
10. To solve
we identify to
= 0, F(s) = (e~t)'
X(t)
= eAtC + eAt
and use the results of Problem 1 and equation (5) in the texto
(t eASF(s) ds
i;
413
Exercises 8.4
11. To solve
= O, F( s) = (~),
we identify to
= (COSht
sinh t
(Cl
Cl
(Cl
Cl
sinh t
+ (COSht
(Cl)
C2
cosh t + C2 sinh t)
sinh t + C2 cosh t
= (Cl cosh t
Cl
sinh t)
cosh t
and use the results of Problem 2 and equation (5) in the texto
sinh t)
cosh t
sinh t
+ (COSht
sinh t)
cosh t
sinh t
+ C2 sinh t) + (COSht
+ C2 cosh t
sinh t
cosh t + C2 sinh t)
sinh t + C2 cosh t
cosh S
(t (
Jo
cosh s
sinh s
' sinh s
 sinh s) (1) ds
cosh s
1
s)
 sinh
coshs
+ (COSht
ds
t
O
sinh t
2
= (Cl cosh t
(t (
Jo
cosh t) = ci (COSht)
cosh2 t
sinh t
+ C2 (Sinh t) _
(1)
cosh t
12. To solve
X' =
we identify to
= O, F(s)
(C~Sh t
sinh t
Cl
( Cl
sinh t)
cosh t
sinh t
1) X
O
+ (COSht )
sinh t
t)
COSh
, and use the results of Problem 2 and equation (5) in the texto
smht
(Cl)
C2
cosh t + C2 sinh t )
sinh t + C2 cosh t
= (Cl cosh t
Cl
=. (
(O1
+ (C~Sh t
sinh t
sinh t)
cosh t
(t (
Jo
cosh s
sinh s
( cosh t sinh t) (t (1 ) ds
sinh t cosh t Jo O
+.
+ C2 sinh t) + (COSht
+ C2 cosh t
sinh t
(Os)
cSlonshhtt)
414
Ito
S)
 sinh
coshs
S)
(C~Sh
sinh s
ds
Exercses 8.4
Cl cosh
= ( Cl sinh
t + C2sinh t )
t + C2cosh t +
Cl cosh
= (
Cl sinh
t + C2sinh t ) + ( t cosh t )
t + C2cosh t
t sinh t
(COSh t
sinh t
t) (t)
t
O
sinh
cosh
= Cl
(COSh t )
sinh t
+ C2 ( sinh t ) + t ( cosh t) .
cosh
13. We have
of the initialvalue
problem is
X=(t:1)_4(t~1)+6(
2t
2t
~ ).
2t + 1
14. We have
X(O) = C3 (~)
+ C4 (~) + (
~3)=
e:~)
= (~) .
Thus, C3 = 7 and C4 = ~, so
x = 7 ( ~ ) et + ~ (~)
S
15. From sI  A =
4'
e2t
+ ( ~ ) .
3 ) we find
s+4
3/2
(sIA)l
1/2
(S2  s+2
1
1
+s2
s+2
and
415
3/4 _ 3/4 )
s2
1/2
s2
s+2
3/2
s+2
sinh
Exercises 8.4
16. From sI _ A
3/2)
e (
1
(S  4
.1
2t
+ e
(1/2)
2t
+ C2
(3/4)
/
1 2
(sIA)l
(S~3 ~ S~2
s3
and
eAt
s2
__2 +_2 )
s3
s2
1
+s3
s2
2e3t _ e2t
 ( e3t _ e2t
_
= (e  C2) (~)
(S 1 5
+ C2
2)
we find
s 1
17. From sI _ A
2t
C3 (~)
e3t + (e
e3t + C4
e)
e2t.
9 ) we find
8+1
(sI  A)l =
_1
s2
(
3
(s2)2
1
(s  2)2
416
+ 2C2)
e)
e2t
(3/4)
/
32
2t
Exercises 8.4
and
18. Frorn sI  A
= (~ s
q (
1+
t
3t) e
2t
+C2
(9t)
1  3t
e 2t .
~12)
we find
(sI  A)
and
At
(e
s+l+l
(s+1)2+1
=
2
(
e
cos t + t sin t
2et sint
(S+1)2+1)
s+l+l
(s+1)2+1
:' sin t
)
:' cos t  :' sin t .
X=e
At
C=
(e
Cl
e
cos t + t sin t
2et sin t
cos t + sin t) t
e
 2 sin t
:' sin t
) (CC21)
:' cos t  et sin t
+ C2 (
sin t
cos t  sin t
417
)t
Exercises 8.4
19. Solving
2'\
3
= 3 and
1
6 ,\
'\2
= 5.
Then
1 1)
P = (1 3 '
1
(3/2
1/2
1/2)
1/2 '
and
D = (~ ~),
so that
PDPl = (
2 1).
3 6
20. Solving
2'\
1
= 1 and
1
2 ,\
'\2
= 3.
Then
P= (1 1)
p_l=(1/21/2)
1 1 '
1/2
1/2 '
and
= (~ ~)
so that
=P
3!
3.
418
Exercises
etD =
f) +
(!
O
A2
O
O)
An
e
+2" O
1 2
O
A~
1
3!
+ t
1 + A2t
+ ~(A2t)2 + ...
O
O
23. From Problems 19, 21, and 22, and equation (1) in the text
X = etAC = PetDp1C
= etAC = PetDp1C
419
JJ
(,3
01 A~
:
jJ+
8.4
Exercses 8.4
25. If det(sI  A) = O, then s is an eigenvalue of A. Thus sI  A has an inverse if s is not an eigenvalue
of A. For the purposes of the discussion in this section, we take s to be larger than the largest
eigenvalue of A. Under this condition sI  A has an inverse.
26. Since A 3
= 0, A is nilpotent. Since
e At = 1 + At
k!
2!
t2
+ A 2  + ... + A ml
2!
tml
(m  1)!
= 0, so
2
eAt = 1+ At + A t; = [~ ~ ~] +
O O 1
t  t2/2
t
[
t  t2/2
1
[=~ ~ ~] t + [ ~ ~ O1]2!_
1
1 1
t + t2/2
~ 1+
1
O 1
.; t2/2
= AX is
Cl]
C2
C3
[Cl(1tt2/2)+C2t+C3(t+t2/2)]
Clt + C2 + C3t
Cl(t_t2/2)+C2t+C3(1+t+t2/2)
t];
soleeExpandjm.c]
Colleet[sol, {el, e2}] / / MatrixForm
The output gives
y(t) =
ci ( _~et
420
X(t) = bl (~)
et
+ b:
C) e
6t
or
x(t)
= 2blet + b2e6t
y(t) = 3blet
If we replace bl with iCl
+ iC2
+ b2t.
+ ~C2, we obtain
matrix exponential.
(b ) x(t) = cle2tcost
(Cl +c2)e2tsint
+ (2Cl + c2)e2t
sin t
+ C3(6e2t + 6et)
= c2(4e2t  3et) + c4(4e2t  4et)
= CI (e2t  et) + C3(_2e2t + 3et)
w(t) = C2(3e2t
=k
( : ), then X'
We see that k
2. Solving for
CI
= O and
= ~.
and
C2
we find
CI =  ~
and
C2 =
i.
3. Since
we see that A = 4 is an eigenvalue with eigenvector K3. The corresponding solution is X = K3t.
4. The other eigenvalue is A2
solution is
2t ) t
( sin 2t)
t
X(t) = CI ( cos
.
e + C2
e.
 sin 2t
cos 2t
421
= (_~).
The general
e)
so that
= (A + 6)(A + 2)
X
Xl =
= ci
= Oso that
e6t +e2
(_~)
C)
e2t.
+ 5 = O. For A = 1 + 2i we obtain
2A
( 1) e
i
(1+2i)t =
(cos 2t) e t
 sin 2t
K = (~) and
+ 2. (sin 2t) e t .
cos 2t
Then
Xl = (
32
= el ( cos 2t)
 sin 2t
2A
+ 2 = O. For A = 1 +
i) e
(l+i)t
sin 2t) t
e.
 cos 2t
+ e? (
we obtain k, = (3 ~
= (3 cos t + sin t) e t + 2. (
2 cos t
i)
and
+ 3 sin t) e.t
2 sin t
 cos t
Then
X
9. We have det(A  Al)
= el (
3 cos t + sin t ) t
e
2 cos t
+ 2)(82 
+ e2
( 
cos t + 3 sin t ) t
e .
2 sin t
+ 3) = Oso that
422
x ~"
en [m ';'v2t  ( ~/2)
~+o, m sinv2t]
e"
H2
v2t+
H3 [ (
=C1
 7)
5 e2t+c2
(COS
...;2t)
...;2sin...;2tj2
v2t] e'
sin
e'
(Sin
...;2t )
...;2cos...;2tj2 et.
et+c3
cos ...;2t
11. We have
x c = c1
(1)
sin ...;2t
e2t + c2
(4)
1
e4t.
Then
cp = (e2t
O
and
U=
cp1 F dt =
cp1 =
4e4t)
e4t '
e2t
(
_4e2t)
<t '
2e2t _ 64te2t )
( 15e2t + 32te2t )
dt =
16te4t
_e4t  4te4t
'
so that
11 + 16t)
Xp = cpU = (
.
1  4t
12. We have
Xc =
Then
cp=
and
2 cos t
(  sin t
1
cp F dt
so that
Xp
Cl
t)
2 cos
(  sin t
t)
2 sin
cos t
+ C2 (
cp1
'
(costsect)
sin t
dt =
2 sin
cost
t) e .
t
t  sin t)
(~C~S
~ sm t
(SintInlsect+tantl)
 cos t
ti)
= cpU = (
.
 1 + sin t In I sec t + tan ti
423
et,
cost
'
cos t + sin t )
2 cos t
+ C2 ( sin t 
cos t )
2 sin t
Then
q,
2 sin t
and
q,lFdt=
U=
J
= (~ cos t
~ cos t 
sin t'
 cos t
!cos t  ~ sin t )
!cos t + ~sin t
'
cot ti)
cot ti
'
so that
Xp = q,U = ( 1)
1
+.(
sin t
) In I ese t  cot ti
sin t + cos t
14. We have
Then
q,l=
t 2t e 2t)
e
e
t 2t
eu
eu
'
and
t1)
( 1
U=Jq,lFdt=J
dt=
(lt2t)
2_t
'
so that
X =q,U=
p
(1/2)t2e2t+
1/2
(2)te2t.
1
= 0, k3 = 1 and then
k2
+ 3k2 + 3k3
= 1, k3 = we get
424
= 0, so kl
(k2
+ k3)'
Choosing
= O we see that
Al
= 3,
two. Letting
K = (~~) ,
k3
as in part (a), we note that (A  OI)K = AK
(k2
+ k3).
Choosing k2
O implies that kl + k2 + k3 = O, so kl
= X = IX.
425
= 3 is
Xl
(t)
CI COS
 2X2
=O
2Xl  (D2
+ 2)X2
= O.
+ 6D2 + 4)XI =
(D2 +4)Xl
+ v5i,
so
)3  v5 t +
C2
sin )3  v5 t +
+ v5 t +
C3 cos )3
C4
sin )3 + v5 t.
Then
1
+ 4)Xl
X2(t) = 2(D2
(t)
1v5
+ 2C3
The frequencies are
WI =
/
r:
cos V 3 + V 5 t
)3  v5 and
W2 =
)3
1v5
2
C4 sin
/
~
V 3  v 5 t.
+ v5.
[4 2] X
2 2
+ v5
and
X = Vsinwt
and 3  v5,
WI
)3  v5 and
W2
VI = [(1 + ~)/2]
and
=CI
[(1+~)/2]
[(1+~)/2]
cosV3v5t+C2
426
sinV3v5t
)3
+ v5.
The
10
A = M1K=
10
A
ol
T
10
10
10
10
2
10
10
Ai
10
10
Wi =
rxi
and
Ti = 21r/ Wi.
3.6420
1.7252
3.5223
1.7838
3.1776
1.9773
3.3641
1.8677
8.8604
7.7256
6.8082
6.2089
1.5000
2.9766
2.1108
2.7795
2.2605
2.6093
2.4080
2.4918
2.5216
1.2247
5.1302
We see that the frequencies are higher and consequently, the periods are smaller.
3. (a) For this exercise we assume a building of four stories, each with mass 10,000 kg and restoring
force constants each k, = 5,000 kg/s2 Then
1 O
M = 10,000
1 O
10,000
5,000 10,000
K=
O O 1 O
O O O 1
5,000
5,000
5,000 10,000
5,000
5,000 5,000
and
1
A=
0.5 O
0.5 1 0.5
O 0.5 1
O
0.5
O
O
0.5
0.5
(b) The table below shows the eigenvalues Ai with the corresponding frequencies Wi =
periods
Ti = 21r/ wl .
A
Ol
T,
1.7660
1.1737
0.5000
0.0603
1.3289
4.7280
1.0834
5.7998
0.7071
8.8858
0.2456
25.5855
427
rxi
and
Chapter
8 Related Exercises
Since the periods are all larger than 4, the building should be safe in a period2 earthquake.
(e) If k is multiplied by 10, then
100
50
A=
50
100
50
50
100
50
50
50
From the table below we see that the periods are generaUy in the range from 1.5 to 3 and the
building is unsafe from a period2 earthquake.
A.
w
T
17.660
11.737
5.000
0.603
4.2024
1.4951
3.4259
1.8340
2.2361
2.8099
0.7766
8.0909
Experimenting, we see that if the building is .made stiffer by a factor of 8, the periods are
entering the danger zone.
A.
w
T
14.128
9.389
4.000
0.482
3.7588
1.67l6
3.0642
2.0505
2.0000
3.1416
0.6946
9.0459
4. The related homogeneous system is X" = M1KX = AX, where A is shown in the solution of
Problem 3(a). We assume solutions of the form X
are 1. 7660, l.1737,
= V cosw and X
0.6565
0.6565
0.2280
0.5774
0.5774
0.4285
0.5774
0.5774
0.2280
V2=
0.5774
0.2280
and
0.4285
V4 =
+ C7V4
cosO.2456t
+ ca V4
sin 0.2456t.
Since
10000cos 3t
O
F(t) =
O
O
428
0.5774
0.6565
Chapter
8 Related Exercises
to solve the nonhomogeneous system we look for a particular solution of the form
bl
al
a2
Xp =
cos3t
a3
a4
+ MIF(t)
9al = 1  al
b2
b3
b4
sin3t.
we obtain
+ 0.5a2
and
0.12549,
a2 =
0.00787,
a3 =
0.00049,
a4
0.00003, and
= b: = b3 = o = O. Thus,
bl
0.12549
X p 
0.00787
cos3t.
0.00049
0.00003
= O,
Cs
is
0.4285
X(t)
0.0592
0.6565
0.5774
0.6565
cos 1.3289t  0.0839
0.2280
0.5774
0.4285
0.2280
0.2280
0.5774
+ 0.0679
0.5774
O
cos 1.0834t
cos 0.0679t
+ 0.0255
0.4285
0.5774
0.6565
0.5774
429
0.12549
cos 0.2456t
0.00787
0.00049
0.00003
cos 3t.
Numerical Solutions of
Ordinary Differential Equations
Exercises 9.1
AH tables in this chapter were constructed in a spreadsheet program which does not support subscripts.
Consequently,
1.
2.
h 
x(n)
1.00
1.10
1.20
1.30
1.40
1.50
h 
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
3.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
Xn
and
0.1
y(n)
5.0000
3.9900
3.2546
2.7236
2.3451
2.0801
0.1
y(n)
2.0000
1.6600
1.4172
1.2541
1.1564
1.1122
0.1
y(n)
0.0000
0.1005
0.2030
0.3098
0.4234
0.5470
Yn
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.05
y(n)
2.0000
1.8150
1.6571
1.5237
1.4124
1.3212
1.2482
1.1916
1.1499
1.1217
1.1056
h = 0.05
x(n)
y(n)
0.00
0.0000
0.05
0.0501
0.10
0.1004
0.15
0.1512
0.20
0.2028
0.25
0.2554
0.30
0.3095
0.35
0.3652
0.40
0.4230
0.45
0.4832
0.50
0.5465
430
Exercses 9. 1
4.
x(n}
0.00
0.10
0.20
0.30
0.40
0.50
5.
0.1
y(n}
0.0000
0.0952
0.1822
0.2622
0.3363
0.4053
0.1
yen}
0.0000
0.0050
0.0200
0.0451
0.0805
0.1266
0.1
yen}
0.5000
0.5215
0.5362
0.5449
0.5490
0.5503
x(n}
0.00
0.10
0.20
0.30
0.40
0.50
6.
x(n}
0.00
0.10
0.20
0.30
0.40
0.50
7.
0.1
y(n}
1.0000
1.1110
1.2515
1.4361
1.6880
2.0488
x(n}
0.00
0.10
0.20
0.30
0.40
0.50
h 
x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
 0.05
x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
y(n}
0.0000
0.0488
0.0953
0.1397
0.1823
0.2231
0.2623
0.3001
0.3364
0.3715
0.4054
0.05
y(n}
0.0000
0.0013
0.0050
0.0113
0.0200
0.0313
0.0451
0.0615
0.0805
0.1022
0.1266
0.05
yen}
0.5000
0.5116
0.5214
0.5294
0.5359
0.5408
0.5444
0.5469
0.5484
0.5492
0.5495
x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
x(n}
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.05
yen}
1.0000
1.0526
1.1113
1.1775
1.2526
1.3388
1.4387
1.5556
1.6939
1.8598
2.0619
431
Exercises 9. 1
8.
9.
10.
h = 0.1
x(n)
y(n)
0.00
1.0000
0.10
1.1079
0.20
1.2337
0.30
1.3806
0.40
1.5529
0.50
1.7557
x(n)
1.00
1.10
1.20
1.30
1.40
1.50
0.1
yen)
1.0000
1.0095
1.0404
1.0967
1.1866
1.3260
h x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.1
y(n}
0.5000
0.5250
0.5498
0.5744
0.5986
0.6224
11.
h
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
h = 0.05
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
x(n)
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1. 35
1.40
1. 45
1. 50
h =
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
y(n)
1.0000
1.0519
1.1079
1.1684
1.2337
1.3043
1.3807
1.4634
1.5530
1.6503
1.7560
0.05
y(n)
1.0000
1.0024
1.0100
1.0228
1.0414
1.0663
1.0984
1. 1389
1.1895
1.2526
1.3315
0.05
y(n}
0.5000
0.5125
0.5250
0.5374
0.5498
0.5622
0.5744
0.5866
0.5987
0.6106
0.6224
0.1
y(n}
2.0000
2.1220
2.3049
2.5858
3.0378
3.8254
exact
2.0000
2.1230
2.3085
2.5958
3.0650
3.9082
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
h  0.05
y(n}
2.0000
2.0553
2.1228
2.2056
2.3075
2.4342
2.5931
2.7953
3.0574
3.4057
3.8840
432
exact
2.0000
2.0554
2.1230
2.2061
2.3085
2.4358
2.5958
2.7997
3.0650
3.4189
3.9082
Exercises
12. (a)
(b)
h=O.l
20
EULER
x{nl
1.00
1.10
1.20
1. 30
1.40
15
10
5
9. 1
IMPROVED
EULER
.v.uu.
_y (nl_
1.0000
1.2000
1.4938
1.9711
2.9060
1.0000
1.2469
1.6668
2.6427
8.7988
~~====;=~~x
1.1 1.2 1.3 1.4
13. (a) Using the Euler method we obtain y(O.l)
~ Yl = 1.2.
(b) Using y" = 4e2x we see that the local truncation error is
Since e2x is an increasing function, e2c :::;e2(0.1) = eO.2 for O :::;e :::;0.1. Thus an upper bound
for the local truncation error is 0.02eO.2 = 0.0244.
(e) Since y(O.l)
~ Y2
1.21.
(e) The error in (d) is 1.2214  1.21 = 0.0114. With global truncation error O(h), when the step
size is halved we expect the error for h
0.1. Comparing
1.22.
= 8e2c
(0.1)3
= 0.001333e2c.
Since e2x is an increasing function, e2c :::;e2(0.1)= eO.2 for O :::;e:::; 0.1. Thus an upper bound
for the local truncation error is 0.001333eo.2 = 0.001628.
(e) Since y(O.l)
= eO.2 =
l.221403,
 Yl
0.001628.
~ Y2 = 1.221025.
when
the step size is halved we expect the error for h = 0.05 to be onefourth the error for h = 0.1.
Comparing 0.000378 with 0.001403 we see that this is the case.
15. (a) Using the Euler method we obtain y(O.l) ~ Yl
0.8.
(b) Using y" = 5e2x we see that the local truncation error is
5e~c (0.1)2
2
= 0.025e2c.
433
Exercises 9. 1
Since e2x is a decreasing function, e2c :S ea = 1 for O:S e :S 0.1. Thus an upper bound for
the local truncation error is 0.025(1)
(e) Since y(O.l)
= 0.025.
= 0.8234, the actual error is y(O.l)  Yl = 0.0234, which is less than 0.025.
(d) Using the Euler method with h = 0.05 we obtain y(O.l) ~ Y2 = 0.8125.
(e) The error in (d) is 0.8234  0.8125 = 0.0109. With global truncation error O(h), when the
step size is halved we expect the error for h
y'" =
= 0.825.
6
Since e2x is a decreasing function, e2c :S ea
and
y"(c) ~2 :S 19(0.1)2(1) = 0.19.
(e) Using the Euler method with h
W(!
obtain
y(1.5) ~ 1.9424.
(d) Since y(1.5)
2.0532, the error for h = 0.1 is Ea.l = 0.2325, while the error for h
0.05 is
Ea.a5 = 0.1109. With global truncation error O(h) we expect Ea. 1/ Ea.a5 ~ 2. We actually have
Ea. 1/ Ea.a5 = 2.10.
18. (a) Using
y'"
= _114e3(xl)
434
Exercses 9. 1
(b) Since e3(xl)
::; e3(11) =
and
lylll(C) ~3 ::; 19(0.1)3(1) = 0.019.
(e) Using the improved Euler method with h
= 0.05
= 2.053216, the
error for h
Eo.os = 0.005950. With global truncation error O(h2) we expect Eo.1/ Eo.os ~ 4. We actuaJly
have Eo.1/ Eo.os = 4.52.
19. ()a
11
1
Usmg y =  (x + 1)2 we see that the local truncation error is
lyll(C) ~21 =
(e:
1)2 ~2.
(b) Since (x ~ 1)2 is a decreasing function for O::; x::; 0.5, (e: 1)2 ::; (O: 1)2 = 1 for O::; e::; 0.5
and
yll(C) ~21 ::; (1) (0.;)2 = 0.005.
(e) Using the Euler method with h = 0.1 we obtain y(0.5) ~ 0.4198. With h
y(0.5) ~ 0.4124.
= 0.05 we obtain
= 0.4055, the error for h = 0.1 is EO.l = 0.0143, while the error for h = 0.05 is
Eo.os = 0.0069. With global truncation error O(h) we expect Eo.1/ Eo.os ~ 2. We actuaJly have
Eo.1/ Eo.os = 2.06.
.
111
2
20. ( a ) U sing y = (x + 1)3 we see that the local truncaton error is
(d ) Since y(0.5)
111
y (e)
h3
6" =
1
h3
(e + 1)3 3"
(b) Since (x ~ 1)3 is adecreasing function for O::; x::; 0.5, (e: 1)3 ::; (O: 1)3 = 1 for O::; e::; 0.5
and
y"I(C) ~3 ::; (1) (0.;)3
(e) Using the improved Euler method with h
= 0.000333.
= 0.405465, the
error for h
= 0.05 is
Eo.os = 0.000046. With global truncation error O(h2) we expect Eo.1/ EO.05 ~ 4. We actually
have Eo.1/ EO.05 = 3.98.
435
Exercises 9. 1
21. Because Y~+l depends on Yn and is used to determine Yn+l, all of the Y~ cannot be computed at one
time independently of the corresponding 'Yn values. For example, the computation of Y4 involves
the value of Y3.
Exercises 9.2
1.
h = 0.1
y(n}
2.0000
2.1230
2.3085
2.5958
3.0649
3.9078
x(nJ
0.00
0.10
0.20
0.30
0.40
0.50
exact
2.0000
2.1230
2.3085
2.5958
3.0650
3.9082
= 2,
method is
Yn+ 1 = Yn  k1
k1
= hf(xn, Yn)
k2 = hf(xn
h = 0.1
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
RK
y(n)
2.0000
2.1205
2.3006
2.5759
3.0152
3.7693
+ 2k2
+ h/4,
Yn
+ kj4).
IMP EULER
y(n)
2.0000
2.1220
2.3049
2.5858
3.0378
3.8254
3.
x(n)
1. 00
1.10
1.20
1.30
1.40
1.50
v(n)
5.0000
3.9724
3.2284
2.6945
2.3163
2.0533
4.
x(n)
0.00
o . 10
0.20
0.30
0.40
0.50
v(n)
2.0000
1.6562
1.4110
1.2465
1.1480
1.1037
5.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
v(n)
0.0000
0.1003
0.2027
0.3093
0.4228
0.5463
6.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
v(n)
1.0000
1.1115
1.2530
1.4397
1.6961
2.0670
7.
x(nJ
0.00
0.10
0.20
0.30
0.40
0.50
v(n)
0.0000
0.0953
0.1823
0.2624
0.3365
0.4055
8.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
v tn)
0.0000
0.0050
0.0200
0.0451
0.0805
0.1266
436
Exercises
9.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
v(n)
0.5000
0.5213
0.5358
0.5443
0.5482
0.5493
ll.
x(n)
l. 00
1.10
1.20
1.30
1.40
1.50
1.0000
1.0101
1.0417
1.0989
1.1905
1.3333
(n)
10.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
y(n)
1.0000
1.1079
1.2337
1.3807
1.5531
1.7561
12.
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
(n)
0.5000
0.5250
0.5498
0.5744
0.5987
0.6225
t n)
0.0
1.0
2.0
3.0
4.0
5.0
v(t)
v(n)
0.0000
25.2570
32.9390
34.9772
35.5503
35.7128
40
30
20
~01~2~34~5t

1 (1
2V32 V32 
JO.025 v
1)
V32 + JO.025
dv
= dt
and
2
vi  ln
1V32 
16v5 (ev'TIt  1)
ev'TIt + 1
437
JO.025 vi) = t
+ c.
9.2
Exercises
9.2
(b) From the graph we estimate A(1) ~ 1.68, A(2) ~ 13.2, A(3) ~ 36.8, A(4) ~ 46.9, and
A(5) ~ 48.9.
A(t)
50
40
30
20
10
1(1
a A + a
(3)
 (3A
1
[In A ln(a
a
dA
 (3A)]
A
In a _ (3A
= dt
= dt
= t +e
= a(t
+ e)
= eo(t+c)
a  (3A
A
= aeo(t+c)
 (3Aeo(t+c)
Thus
aeo(t+c)
a
a
 ....,;
 1+ (3eo(t+c)  (3 + eo(t+c)  (3 + eoceot
A(t) 
eoc =
438
Exercises 9.2
A( ) ~
2.128
t
A
A
A
(da s)
(observed)
(approximated)
(exact)
15. (a)
h  0.05
x(n)
y(n)
1.00
1.05
1.10
1.15
1.20
1.25
1. 30
1.35
1.40
1.0000
1.1112
1.2511
1.4348
1.6934
2.1047
2.9560
7.8981
1.06E+15
1
2.78
1.93
1.95
2
13.53
12.50
12.64
3
36.30
36.46
36.63
4
47.50
47.23
47.32
5
49.40
49.00
49.02
(b)
h  0.1
v(n)
1.0000
20
1.2511
15
1.6934
10
2.9425
~========~x
903.0282
1.1
(b) Using
y(5)
(x)
;::::Yl
1.2
1.3
1.4
1.2214.
120
= 32e2c (0.1)5
120
= 0.000002667e2c.
Since e2x is an increasing function, e2c :S e2(O.1) = eO.2 for O :S e :S 0.1. Thus an upper bound
for the local truncation error is 0.000002667eo.2
(e) Since y(O.l)
= eO.2
= 0.000003257.
than 0.000003257.
(d) Using the Iourthorder RungeKutta
;::::Y2 =
1.221402571.
(e) The error in (d) is 1.221402758  1.221402571 = 0.000000187. With global truncation error
O(h4), when the step size is halved we expect the error for h = 0.05 to be onesixteenth the
= 0.1. Comparing 0.000000187 with 0.000002758 we see that this is the case.
(a) Using the fourthorder RungeKutta method we obtain y(O.l) ;::::Yl = 0.823416667.
(b) Using y(5) (x) = 40e2x we see that the local truncation error is
error for h
17.
eO
= 0.000003333.
439
= 1 for O :S e
Exercises
9.2
(e) Since y(O.l) = 0.823413441, the actual error is ly(O.l)  Yll = 0.000003225, which is less than
0.000003333.
(d) Using the fourthorder RungeKutta method with h
~ Y2
0.823413627.
(e) The error in (d) is 10.823413441 0.8234136271 = 0.000000185. With global truncation error
O(h4),
error when h
= 0.1. Comparing 0.000000185 with 0.000003225 we see that this is the case.
l
(b) Since e3(xl)
y(5)(C).!i!_1 = 8.55h5e3(cl).
120
and
y(5) (c) 1~50:S 8.55(0.1)5 (1) = 0.0000855.
(e) Using the fourthorder RungeKutta method with h = 0.1 we obtain y(1.5) ~ 2.053338827.
With h = 0.05 we obtain y(1.5) ~ 2.053222989.
.
()
24
19. (a) Usmg y 5 =
we see that the local truncation error is
(x + 1)5
(5)
y
(b) Since (x
+ 1)5 is a decreasing
c .!i!_ _
1
h5
() 120  (c + 1)5 5
1
1
function for O :S x :S 0.5, (c + 1)5 :S (O + 1)5 = 1 for O :S c:S 0,5
and
y(5)(C) ~5 :S (1) (0.;)5
= 0.000002.
(e) Using the fourthorder RungeKutta method with h = 0.1 we obtain y(0.5) ~ 0.405465168.
With h = 0.05 we obtain y(0.5) ~ 0.405465111.
20. Each step of the Euler method requires only 1 function evaluation, while each step of the improved
Euler method requires 2 function evaluations  once at (xn, Yn) and again at (Xn+l, y~+l)'
The
secondorder RungeKutta methods require 2 function evaluations per step, while the fourthorder
RungeKutta method requires 4 function evaluations per step. To compare the methods we approximate the solution of y' = (x
+y
method, h = 0.05 for the improved Euler method, and h = 0.025 for the Euler method. For each
method a total of 8 function evaluations is required. By comparing with the exact solution we see
that the fourthorder RungeKutta method appears to still give the most accurate resulto
440
Exercises
EULER
IMP EUL
RK
0.025 h = 0.05 h = 0.1
y(n)
y_(n)
y(n)
2.0000
2.0000
2.0250
2.0526
2.0553
2.0830
2.1165
2.1228
2.1230
2.1535
2.1943
2.2056
2.2395
2.2895
2.3075
2.3085
9.2
h
x(n)
0.000
0.025
0.050
0.075
0.100
0.125
0.150
0.175
0.200
exact
2.0000
2.0263
2.0554
2.0875
2.1230
2.1624
2.2061
2.2546
2.3085
y
+y =
d
dx [eXy] = 10ex sin 3x ~
eXy
~
When x
3 cos 3x
= O, y
= O, so O = 3
+ 3ex.
= eX sin 3x
+e
 3ex cos 3x
y = sin 3x  3 cos 3x
+ c and c =
+ cex.
y(n)
0.0000
0.1440
0.5448
1.1409
1.8559
2.6049
3.3019
3.8675
4.2356
4.3593
4.2147
x(n)
1.0
1.1
1.2
1.3
1.4
1.5
1'.6
1.7
1.8
1.9
2.0
= 1.53236.
y
y(n)
4.2147
3.8033
3.1513
2.3076
1.3390
0.3243
0.6530
1.5117
2.1809
2.6061
2.7539
441
Exercises 9.3
Exercises 9.3
~[eXyl
dx
= (x
 l)eX
and
y
=e
X(
xex
+ e) =
x
+ ce".
From y(O) = 1 we find e = 1 and y = x + e". Comparing exact values with approximations
obtained in Example 1, we find y(0.2)
+K1j2)
+K2j2)
310 K4=H*FNF(X+H,Y+K3)
320 Y=Y+(Kl+2*K2+2*K3+K4)j6
330 Z(I+l)+ Y
442
Exercises
9.3
340 X=X+H
350 PRINT X,Y
360 NEXT 1
370 REM COMPUTE REMA1N1NG X AND Y VALUES
380 FOR 1=4 TO N
390 YP= Y+ H*(55*FNF(X,Z( 4))59*FNF(XH,Z(3) )+37*FNF(X2*H,Z(2))
9*FNF(X3*H,Z(1)) )/24
400 Y= Y+ H* (9*FNF(X + H,YP)+ 19*FNF(X,Z( 4))5*FNF(XH,Z(3))+ FNF(X2*H,Z(2))) /24
410 X=X+H
420 PRINT X,Y
430 Z(1)=Z(2)
440 Z(2)=Z(3)
450 Z(3)=Z(4)
460 Z(4)=Y
470 NEXT 1
480 END
3.
x (n)
0.00
0.20
0.40
0.60
0.80
4.
x iri)
0.00
0.20
0.40
0.60
0.80
y(n)
1.0000
0.7328
0.6461
0.6585
0.7332
0.7232
y(n)
2.0000
1.4112
1.1483
1.1039
1.2109
1.2049
initia1 condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
initia1 condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
443
Exercises
5.
x(n)
0.00
0.20
0.40
0.60
0.80
1.00
9.3
y(n)
0.0000
0.2027
0.4228
0.6841
1.0234
1.0297
1.5376
1.5569
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
6.
x(n)
0.00
0.20
0.40
0.60
0.80
1.00
y(n)
1.0000
1.4414
1.9719
2.6028
3.3483
3.3486
4.2276
4.2280
x (n)
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
444
'y(n)
0.0000
0.1003
0.2027
0.3093
0.4227
0.4228
0.5462
0.5463
0.6840
0.6842
0.8420
0.8423
1.0292
1.0297
1.2592
1.2603
1.5555
1.5576
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
y(n)
1.0000
1.2102
1.4414
1.6949
1.9719
1.9719
2.2740
2.2740
2.6028
2.6028
2.9603
2.9603
3.3486
3.3486
3.7703
3.7703
4.2280
4.2280
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
Exercises 9.3
7.
x(n)
0.00
0.20
0.40
0.60
0.80
1. 00
y(n)
0.0000
0.0026
0.0201
0.0630
0.1362
0.1360
0.2379
0.2385
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1. 00
8.
x tn)
0.00
0.20
0.40
0.60
0.80
1.00
y(n)
1.0000
1.2337
1.5531
1.9961
2.6180
2.6214
3.5151
3.5208
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
x(n)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
445
y(n)
0.0000
0.0003
0.0026
0.0087
0.0201
0.0200
0.0379
0.0379
0.0630
0.0629
0.0956
0.0956
0.1359
0.1360
0.1837
0.1837
0.2384
0.2384
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
v tn)
1.0000
1.1079
1.2337
1.3807
1.5530
1.5531
1.7560
1.7561
1.9960
1.9961
2.2811
2.2812
2.6211
2.6213
3.0289
3.0291
3.5203
3.5207
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
predictor
corrector
Exercises 9.4
Exercises 9.4
1. The substitution
y'
= u leads to the
Yn+l = Yn
= 2
_
iteration formulas
+ hun,
Un+l = Un
+ h(4un
and Uo = 1. Then
u. = Yo + O.luo
= 2
+ 0.1(1) =
1.9
+ 0.1(2.2)
= 2 and
C2 = 5.
 4Yn).
= 1.68.
= _2e2x + 5xe2x
Thus y
+ 8) = 2.2
+ hun,
Yn+l = Yn
The initial conditions are yO
=4
Un+
1 = un + h (~ Un  :2 Yn) .
and Uo = 9. Then
Yl
= yo + O.luo
= 4 + 0.1(9)
Ul
(2
Y2 = Yl
+ O.lUl
= 4.9
2)
= 4.9
+ 0.1(10) = 5.9.
C2 = 5.
= 10
Thus y
C1X
+ c2x2.
= 6.
u'
= 4u  4y.
Using formula (4) in the text with x corresponding to t, y corresponding to x, and u corresponding
to y, we obtain
RUllfLeKuttamethod with h=0.2
ml
m2
m3
m4
kl
k2
k3
k4
0.2000
0.4400
0.5280
0.9072
2.4000
3.2800
3.5360
4.8064
ml
m2
m3
m4
kl
k2
k3
k4
0.1000
0.2443
0.1600
0.3298
0.1710
0.3444
0.2452
0.4487
1.2000
1.7099
1.4200
2.0031
1.4520
2.0446
1.7124
2.3900
RungeKutta
446
0.00 2.0000
0.20 1.4928
1.0000
4.4731
Exercises
4. The substitution y'
9.4
y' = u,
x2
Using formula (4) in the text with x corresponding to t, y corresponding to x, and u corresponding
to
y,
we obtain
RungeKutta
ml
m2
m3
m4
kl
k2
k3
k4
1.8000
2.0000
2.0017
2.1973
2.0000
2.0165
1.9865
1.9950
1.00
1.20
4.0000
6.0001
9.0000
11.0002
RungeKutta
ml
m2
m3
m4
kl
k2
k3
k4
0.9000
1.0000
0.9500
1.0500
0.9501
1.0501
0.9998
1.0998
1.0000
1.0000
1.0023
1.0019
0.9979
0.9983
0.9996
0.9997
1.00
1.10
1.20
4.0000
4.9500
6.0000
9.0000
10.0000
11.0000
y' = u,
Using formula (4) in the text with
= 2u  2y + etcost.
ml
0.4000
m2
0.4600
m3
0.4660
m4
0.5320
kl
0.6000
k2
0.6599
k3
k4
0.6599
ml
0.2000
0.2315
m2
0.2150
0.2480
m3
0.2157
0.2487
m4
0.2315
0.2659
kl
k2
0.3000
0.3299
0.3150
0.3446
kl
k2
k3
0.3150
0.3444
6.
ml
10.0000
8.7500
10.1563
13.2617
17.9712
m2
m3
0.0000 12.5000
2.500013.4375
4.375017.0703
6.3672 22.9443
8.8867 31.3507
m4
0.0000
20.0000
28.7500 5.0000
40.0000 8.7500
55.1758 12.7344
75.9326 17.7734
5.0000
4.3750
5.0781
6.6309
8.9856
447
k4
k3
0.3298
0.3587
we obtain
0.00
O .7170 0.20
RungeKutta
y,
1.0000
1.4640
2.0000
2.6594
0.00 1.0000
0.10 1.2155
0.20 1.4640
u
2.0000
2.3150
2.6594
Exercises 9.4
i1,i2
7
i1
i2
2
1
As t ;
00
5t
7.
RungeKutta
ml
m2
m3
m4
kl
k2
k3
k4
2.0000
2.2800
2.3160
2.6408
1.2000
1.4000
1.4280
1.6632
0.00
0.20
6.0000
8.3055
2.0000
3.4199
ml
m2
m3
m4
kl
k2
k3
k4
1.0000
1.1494
1.0700
1.2289
1.0745
1.2340
1.1496
1.3193
0.6000
0.7073
0.6500
0.7648
0.6535
0.7688
0.7075
0.8307
RungeKutta
x,y
0.511.52t
448
0.00 6.0000
0.10 7.0731
0.20 8.3055
y
2.0000
2.6524
3.4199
Exercises
8.
RungeKutta
ml
m2
m3
m4
0.6000
0.9400
1.1060
1.7788
kl
k2
k3
k4
0.00 1.0000
1.40002.06002.39403.72120.202.07853.3382
RunqeKutta
ml
m2
m3
m4
kl
k2
k3
k4
0.3000
0.5193
0.3850
0.6582
0.4058
0.6925
0.5219
0.8828
0.7000
1.1291
0.8650
1.4024
0.9068
1.4711
9.4
1.0000
0.00 1.0000
1.1343 0.10 1.4006
1.84740.202.0845
1.0000
1.8963
3.3502
x,y
50
40 y(t)
30
9.
RungeKutta
m1
m2
m3
m4
kl
k2
k3
k4
0.00 3.0000
1.0000 0.9200 0.9080 0.8176 0.6000 0.7200 0.7120 0.8216 0.20 3.9123
RungeKutta
m1
m2
m3
m4
kl
k2
k3
k4
5.0000
4.2857
0.00 3.0000
0.5000 0.4800 0.4785 0.4571 0.3000 0.3300 0.3290 0.3579 0.10 3.4790
0.4571 0.4342 0.4328 0.4086 0.3579 0.3858 0.3846 0.4112 0.20 3.9123
449
5.0000
4.6707
4.2857
Exercises
9.4
x,y
10.
RunqeKutta
ml
m2
m3
m4
kl
k2
k3
k4
0.6400
1.2760
1.7028
3.3558
1.3200
1.7720
2.1620
3.5794
RungeKutta
ml
m2
m3
m4
kl
k2
k3
k4
0.3200
0.7736
0.4790
1.0862
0.5324
1.1929
0.7816
1.6862
0.6600
1.0117
0.7730
1.2682
0.8218
1.3692
1.0195
1.7996
x,y
2
1.5
0.1
0.2
0.3t
+ y + 5t
x'
y'
= 2x + y  2t.
2x
450
0.00 0.5000
0.20 2.1589
y
0.2000
2.3279
0.00 0.5000
0.10 1.0207
0.20 2.1904
y
0.2000
1.0115
2.3592
Exercses
RungeKutta
mI
m2
m3
m4
kl
0.0000
k2
k3
k4
mI
m2
m3
kl
m4
0.00
0.2000 0.1680 0.3584 0.20
RungeKutta
k2
k3
k4
9.4
1.0000
0.4179
2.0000
2.1824
0.00
0.4000 0.3350 0.3440 0.2858 0.0000 0.0500 0.0460 0.0934 0.10
0.2866 0.2376 0.2447 0.2010 0.0929 0.1362 0.1335 0.1752 0.20
1.0000
0.6594
0.4173
2.0000
2.0476
2.1821
x,y
X'
4 t
= y  3t
2
+ 2t 
RungeKutta
mI
m2
m3
m4
kl
k2
k3
k4
1.1000
1.0910
1.0915
1.0949
RungeKutta
mI
m2
m3
m4
0.00
0.20
3.0000
1.9867
1.0000
0.0933
kl
k2
k3
k4
0.5500
0.5456
0.5470
0.5457
0.5471
0.5457
0.5456
0.5473
0.00
0.10
0.20
3.0000
2.4727
1.9867
1.0000
0.4527
0.0933
451
Exercises 9.4
x,y
60
40
y(t)
20
4
20
8 t
x(t)
40
60
Exercises 9.5
1. We identify P(x)
= o, Q(x) = 9, f(x)
= O, and h
(2  0)/4
equation is
Yi+l
+ 0.25Yi + Yil = O.
0.5
5.6774
2. We identify P(x)
1.0
2.5807
1.5
6.3226
2.0
1.0000
0)/4
equation is
+ Yil
Yi+l  2.0625Yi
= 0.0625x;'
0.25
0.0172
3. We identify P(x)
0.50
0.0316
2, Q(x)
0.75
0.0324
1. 00
0.0000
0)/5
equation is
1.2Yi+l  1.96Yi
+ 0.8Yil = 0.04(5xi).
0.0
0.0000
0.2
0.2259
0.4
0.3356
0.6
0.3308
0.8
0.2167
1.0
0.0000
452
Exercises
9.5
4. We identify P(x) = 10, Q(x) = 25, f(x) = 1, and h = (1  0)/5 = 0.2. Then the finite difference
equation is
Yi
+ 2Yil = 0.04.
0.2
1.9600
0.4
3.8800
0.6
7.7200
0.8
15.4000
1.0
0.0000
5. We identify P(x) = 4, Q(x) = 4, f(x) = (1 + x)e2x, and h = (1  0)/6 = 0.1667. Then the finite
difference equation is
0.6667Yi+l  1.8889Yi+ 1.3333Yi_l = 0.2778(1
+ xi)e2xi.
0.0000
3.0000
0.1667
3.3751
0.3333
3.6306
0.5000
3.6448
0.6667
3.2355
0.8333
2.1411
1.0000
0.0000
difference equation is
1.4167Yi+l  2Yi
+ 0.5833Yi_l = 0.2778(4.JXi).
1.0000
1.0000
= O, and
difference equation is
0.1875)
(
0.0469)
(
0.1875)
Yi+l+ 2+2 Yi+ 1. Yil=O.
( 1+.X,
xi
x,
The solution of the corresponding linear system gives
1.000
5.0000
1.125
3.8842
1.250
2.9640
1.375
2.2064
1.500
1.5826
1.625
1.0681
1.750
0.6430
1.875
0.2913
2.000
0.0000
difference equation is
0.0625)
(
 Yi+l + 2
( 1  o
xt
0.0156)
+ 2
xi
(
0.0625)
Yi + 1 + o  Yil
xt
0.0156lnxi.
1.125
1.250
1.375
1.500
1.625
1.750
1.875
2.000
0.1988 0.4168 0.6510 0.8992 1.1594 1.4304 1.7109 2.0000
453
Exercises 9.5
9. We identify P(x)
= 1x,
Q(x)
= x, f(x)
= (10)/10
= x, and h
equation is
[1 + 0.05(1  Xi)]Yi+l
+ [2 + O.OlXi]Yi + [1 
0.05(1  Xi)]Yil
= O.Olxi.
0.1
0.2660
0.2
0.5097
0.3
0.7357
0.4
0.9471
I
10. We identify P(x)
0.5
1.1465
0.6
1.3353
0.7
1.5149
0.8
1.6855
0.9
1.8474
1.O
2.0000
equation is
0.1
0.8929
0.2
0.7789
0.3
0.6615
0.4
0.5440
0.5
0.4296
0.6
0.3216
0.7
: 0.2225
0.8
0.1347
= O, and h = (1  0)/8
Yi+l 
2.0625Yi
0.9
0.0601
1.O
O .0000 :
+ Yil = O.
0.000
0.0000
0.125
0.3492
0.250
0.7202
0.375
1.1363
0.500
1.6233
0.625
2.2118
= O, and h
0.750
2.9386
= (4  1)/6
0.875
3.8490
1.000
5.0000
equation is
0.5)
( 1 + G Ui+l  2Ui
+ (0.5)
1 G
Uil
= O.
1.0
2.0
3.0
3.5
4.0
1.5
2.5
50.0000 72.2222 83.3333 90.0000 94.4444 97.6190 100.0000
(1 +
%Pi)
Yi+l
+ (2 + h2Qi)Yi +
(1 
%Pi)
Yil
= h2fi
is the same as the one derived on page 434 in the texto The equations are the same because
the derivation was based only on the differential equation, not the boundary conditions. If we
454
o to
= y(O 
h), and Yl
Yl  Yl = 2h.
(1+
~PO) Yl
+ (2 + h2Qo)YO +
(1
(1 +
~Po) Yl
+ (2 + h2Qo)YO +
(1  ~po)
(Yl  2h)
h2 fo
or
2m + (2
+ h2Qo)yo
= h2 fo
ti
+ 2h 
Po.
= 5 we obtain
0.0
0.2
0.4
0.6
0.8
1.0
2.2755 2.0755 1.8589 1.6126 1.3275 1.0000
14. Using h = 0.1 and, after shooting a few times, y'(O) = 0.43535 we obtain the following table with
the fourthorder RungeKutta method.
0.0
1.00000
0.1
1.04561
0.2
1.09492
0.3
1.14714
0.4
1.20131
0.5
1.25633
0.6
1.31096
0.7
1.36392
0.8
1.41388
0.9
1.45962
EULER
2.0000
2.1386
2.3097
2.5136
2.7504
3.0201
IMPROVED
EULER
2.0000
2.1549
2.3439
2.5672
2.8246
3.1157
RUNGE
KUTTA
2.0000
2.1556
2.3454
2.5695
2.8278
3.1197
h=0.05
x(n)
1. 00
1. 05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
EULER
2.0000
2.0693
2.1469
2.2328
2.3272
2.4299
2.5409
2.6604
2.7883
2.9245
3.0690
455
IMPROVED
EULER
2.0000
2.0735
2.1554
2.2459
2.3450
2.4527
2.5689
2.6937
2.8269
2.9686
3.1187
1.0
1.50003
_
RUNGE
KUTTA
2.0000
2.0736
2.1556
2.2462
2.3454
2.4532
2.5695
2.6944
2.8278
2.9696
3.1197
3.
.h=iU.
x(n)
0.50
0.60
0.70
0.80
0.90
1.00
4.
EULER
0.0000
0.1000
0.2010
0.3049
0.4135
0.5279
IMPROVED
EULER
0.0000
0.1005
0.2030
0.3092
0.4207
0.5382
RUNGE
KUTTA
0.0000
0.1003
0.2026
0.3087
0.4201
0.5376
h0.05
x(n)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
EULER
0.0000
0.0500
0.1001
0.1506
0.2017
0.2537
0.3067
0.3610
0.4167
0.4739
0.5327
IMPROVED
EULER
0.0000
0.0501
0.1004
0.1512
0.2027
0.2552
0.3088
0.3638
0.4202
0.4782
0.5378
RUNGE
KUTTA
0.0000
0.0500
0.1003
0.1511
0.2026
0.2551
0.3087
0.3637
0.4201
0.4781
0.5376
EULER
0.5000
0.6000
0.7095
0.8283
0.9559
1.0921
IMPROVED
EULER
0.5000
0.6048
0.7191
0.8427
0.9752
1.1163
RUNGE
KUTTA
0.5000
0.6049
0.7194
0.8431
0.9757
1.1169
h=0.05
x(n)
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
EULER
0.5000
0.5500
0.6024
0.6573
0.7144
0.7739
0.8356
0.8996
0.9657
1.0340
1.1044
IMPROVED
EULER
0.5000
0.5512
0.6049
0.6609
0.7193
0.7800
0.8430
0.9082
0.9755
1.0451
1.1168
RUNGE
KUTTA
0.5000
0.5512
0.6049
0.6610
0.7194
0.7801
0.8431
0.9083
0.9757
1.0452
1.1169
EULER
1.0000
1.2000
1.4760
1.8710
2.4643
3.4165
IMPROVED
EULER
1.0000
1.2380
1.5910
2.1524
3.1458
5.2510
RUNGE
KUTTA
1.0000
1.2415
1.6036
2.1909
3.2745
5.8338
h=0.05
x(n)
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35
1.40
1.45
1.50
EULER
1.0000
1.1000
1.2183
1.3595
1.5300
1.7389
1.9988
2.3284
2.7567
3.3296
4.1253
IMPROVED
EULER
1.0000
1.1091
1.2405
1.4010
1.6001
1.8523
2.1799
2.6197
3.2360
4.1528
5.6404
RUNGE
KUTTA
1.0000
1.1095
1.2415
1.4029
1.6036
1.8586
2.1911
2.6401
3.2755
4.2363
5.8446
1EQ_J.
x(n)
1.00
1.10
1.20
1.30
1.40
1.50
5. Using
Yn+l = Yn
+ hun,
YO = 3
Uo
= 1
we obtain (when h = 0.2) Yl = y(0.2) = YO + huo = 3 + (0.2)1 = 3.2. When h = 0.1 we have
Yl
Ul = Uo
Y2
Yl
+ O.lUl =
456
x(n)
y(n)
0.00
0.10
0.20
0.30
0.40
7. Using Xo
2.0000
2.4734
3.1781
4.3925
6.7689
7.0783
initial condition
RungeKutta
RungeKutta
RungeKutta
predictor
corrector
YI = YO
+ h(xo 
X2 =
+ h(XI + YI) =
YO)
= 2 + 0.1(1  2) = 1.9
and
Xl
+ h(XI
Y2 = YI
1.3 + 0.1(1.3
+ 1.9) =
1.62
= 1.84.
+ x),
= 1, and h
f(x)
= (1  0)/10
difference equation is
Yi+l
or
Yi+l
0.1
4.1987
0.2
8.1049
0.5
0.4
0.3
11.3840 13.7038 14.7770
0.7
12.5396
457
0.6
14.4083
0.8
9.2847
0.9
4.9450
1. O
0.0000
10
If (x,y)
is a critical point, y =
= y,
y' = 9sinx.
and 9sinx
o.
x' = y,
If (x,y)
and so 2x
 y2
critical point.
3. The corresponding plane autonomous system is
X' = y,
If (x, y) is a critical point, y
point.
4. The corresponding plane autonomous system is
I
X = y,
y = 4
1 + x2
x
and so 4 2
l+x
 2y.
2(0) = O. Therefore x
and so x = 0,
X = y,
+ EX3 .
If (x,y)
is a critical point, y =
#.
and x
y = x
and (#,0).
y' = x
458
+ Exlxl.
+ Ex2) =
#,
Exercises
+ Exlxl = x( 1 + Elxl) = O.
+ xy =
into y  xy
+ y) =
O we have x(l
= O, we obtain
Hence x
10. 1
= O, l/E, l/E.
The
O).
= O or
O. Therefore x
y = 1. If x = O, then,
= O, Likewise, if y = 1, 1 + x = O or
substituting
that (O,O) and (1, 1) are critical points of the system.
8. From y2  x
= O we have x
= O, we obtain y4 y
= O or y(y3 1) = O.
It fo11owsthat y = O, 1 and so (O, O) and (1,1) are the critical points of the system.
9. From x  y = O we have y
= x. Substituting into
3x2  4y = O we obtain
3x2  4x
= x(3x
 4)
= O.
It fo11owsthat (O,O) and (4/3,4/3) are the critical points of the system.
10. From x3  y = O we have y = x3. Substituting into x  y3 = O we obtain x  x9 = O or x(l  x8).
Therefore x
= O, 1, 1 and so the critical points of the system are (O, O), (1,1), and (1, 1).
+ h = 10.
= O or
+y =
h + 6)
= O, y
= O or y
= 16. If x + h = lO, we
= O. Since
eX  1
eO 
arbitrary value of y. Therefore any point of the form (O,y) is a critical point.
14. From sin y
= Owe have
y = mL From eXY
= O or x
= y. The
critical points of the system are therefore (mr, mr) for n = O, 1,2, ., ..
15. From x(l  x2  3y2) = O we have x = O or x2
+ 3y2 =
y(3  x2  3y2) gives y(3  3y2) = O. Therefore y = O, 1, 1. Likewise x2 = 1  3y2 yields 2y = O
so that y
O and x2
1  3(0)2
1. The critical points of the system are therefore (O, O), (0,1),
= O or x
= 1, 1. Finally y
= 2 yields
or x = 1, 1. The critical points of the system are therefore (O,O), (1,2), (1,2),
8(1x2)
=O
(1, 2).
17. (a) From Exercises 8.2, Problem 1, x = clt
Cl
= O and
459
C2
+ c2et.
= 2. Therefore x = 2et
and y = 2et.
Exercises 10. 1
(e)
y
(2,2)
2
~
2
2
18. (a) From Exercises 8.2, Problem 6, x = Cl + 2C2e5t and y = 3Cl + C2e5t.
(b ) From X(O)
(e)
= 1+
4
2
4
l. Therefore x
2
/(3,4)
4 x
/~
~
~,4
Cl(5 cos 3t) + C2(5 sin 3t). Al! solutions are one periodic with p = 27f/3.
(b) From X( O) = (4, 5) it follows that Cl
y = 5 cos 3t.
(e)
6
(4,5)
6 x
6
6
20. (a) From Exercises 8.2, Problem 34, x = Cl(sin t  cos t) + C2( cos t  sin t) and y = 2Clcos t
2C2sin t. Al! solutions are periodic with p = 27f.
(b ) Frorn X(O)
l. Therefore x
460
Exercises
(e)
10. 1
y
4
4
4 x
4
+ 2c2(sin t) t.
+ C2(
and y
(e)
,s
5
5
C2 =
y
2
(2,1)
2
+ c2et(2cos2t
+ 2sin2t)
and
(e)
 2sin2t)
2
461
Exercises
10. 1
dt
+ y dY)
dt
dt
de
1 (
dx
dt = r2 y dt
+x
1 xy  x 2 T 4
= (
r
dY)
1 2
dt = r2 (y
+ xy
 y 2 T 4 ) = ==T 5
+ xyr 4 + x 2 
xyr ) = 1.
dr
5
If we use separation of variables on  = ==T we obtain
dt
T
= ( 4t
+ CI
)1/4
and
e=t+C2.
C2
= O and
C1
1
= 256' The final
r=
e=t
. Y1024t + 1 '
and so the solution spirals toward the origin as t increases.
24. Switching to polar coordinates,
dr = 1 (dX
x
dt
dt
+ y dY)
dt
de
1 (
dx
dt = r2 y dt
+x
= 1(xy  x 2 r 2  xy
dY)
1
dt = r2 (y
 xyr
+ y 2r 2 ) = r 3
 x
+ xyr 2 ) = 1.
e = t + C2.
C2
= O and
C1
The final
r
Note that r
+
00 as t + (3\) .
e = t.
= vr.;1==3;;';2~t
'
l2' the
dr
1 (dX
dY)
1
2
2
dt=;
Xdt+Ydt
=;[xy+x(lr)+xy+y
de
1 (
dx
dt = r2 y dt
+x
dY)
1 2
2
2
dt = r2 [y  xy(l  r ) + x
(lr
+ xy(l
)]=r(lr)
 r )] = 1.
462
Exercises
= 1 + cle2t and so r2 =
1
1 + Cle
10. 1
e = t + C2.
r = v!:I:=+=c=l=e ==2~t
'
If X(O) = (1, O), r = 1 and
y
= r sin t
= O. Therefore
e = O when t = O. Therefore
= 3/4,
C2
= O and so
r=
e = t.
VI  ~ e2t
dr = 1 (dX
x  + y dY)
dt
de
dt
dt
1 (
dx
r2 y dt
dt
+x
= 1 [xy  x2
dY)
1 [
dt = r2 y
(4 
T ) 
xY
+ 7(4

xy  y2 (4 r
2 ]
T )
2
2
xY
2
r )  x  7(4  r)
=1.
cle4t
= 2 11 +
 ele4t
and
= 1 and e = O when t = O. It
r
=2
1 le4t
1+
3
~e4t
e = t + C2.
follows that
and
e2
= O and
el
= ~.
Therefore
e = t.
that the solution generates the cirele r = 2 traversed in the elockwise direction. Note also that the
original system is not defined at (O,O)but the corresponding polar system is defined for r = O. If the
RungeKutta method is applied to the original system, the solution corresponding to X(O)
will stall at the origino
27. The system has no critical points, so there are no periodic solutions.
28. Frorn x(6yl)
= Oand y(28x)
are
critical points. From the graph we see that there are periodic solutions
around (1/4,1/6).
463
= (1, O)
Exercises 10. 1
29. The only critical point is (O,O). There appears to be a single periodic
solution around (O,O).
30. The systern has no critical points, so there are no periodic solutions.
31. If X(t)
= (x(t),y(t))
is a solution,
d
of dx
of dy
 f(x(t) y(t)) =   +   = QP  PQ = O
dt
'
ox dt
ay dt
'
using the chain rule. Therefore f(x(t),y(t))
curve of f.
Exercises 10.2
1. (a) If X(O)
= Xo lies on the line y = 2x, then X(t) approaches (O, O) along this lineo For all other
initial conditions, X(t) approaches (O,O) frorn the direction determined by the line y = x/2.
(b)
'
<;
<,
<,
'1
,
<,
'
1
2. (a) If X(O)
'Y = x/2
= Xo lies on the line y = x, then X(t) becornes unbounded along this lineo For all
other initial conditions, X(t) becornes unbounded and y = 3x/2 serves as an asyrnptote.
464
Exercses 10.2
(b)
,,
\
,,
,
4
(1,1)
2
2
4 x
2
\
v
\'yy3XI21
\,
4
(1,1)
5
5
1
1
5. (a) AH solutions approach (O,O) from the direction specified by the line y = X.
465
Exercises
10.2
(b)
1
1
4
3
4 3 ___
2 t l"1
4 x
/':')XI2' ~:
4
7. (a) If X(O)
= Xo
= 3x,
then X(t) approaches (O,O) along this lineo For all other
3
2
3 2 1
/
1
1
2
3
466
Exercses 10.2
(b)
3
2
(1,1)
3
3 x
2
3
< 0, we may conclude from Figure 10.18 that (O,O) is a saddle point.
9. Since b. = 41
= 12, 72  4b. > and so from Figure 10.18, (O,O) is a stable node.
< 0, we may conclude from Figure 10.18 that (0,0) is a saddle point.
= 3 and so from Figure 10.18, (O,O) is a stable spral point.
= 1 and 7 = 2, 72 
4b.
= 1
IJ.LI
< 1.
18. Note that b. = 1 and 7 = J.L. Therefore we need both 7 = J.L < and 72  4b. = J.L2  4 < for (O,O)
to be a stable spiral point. These two conditions may be written as 2 < J.L < O.
19. Note that b. = J.L + 1 and 7 = J.L + 1 and so 72  4b. = (J.L + 1) 2  4 (J.L + 1) = (J.L + 1) (J.L  3). 1t follows
20. 7
2a, b.
<
if and only if 1
<
J.L
a2 + (32 > 0, and 72  4b. = 4(3 < O. If a < 0, (O,O) is a stable spiral point. If
a > 0, (O,O) is an unstable spiral point. Therefore (O,O) cannot be a node or saddle point.
F or X = AIF. Since Xp(t) = AIF is a particular
solution, it follows from Theorem 8.6 that X(t) = Xc(t)+XI is the general solution to X' = AX+F.
If 7 < and b. > then Xc(t) approaches (O,O)by Theorem 10.1(a). It follows that X(t) approaches
X as t _, oo.
22. If be < 1, b. = adxy(l
stable node.
 be)
>
and 72  4b.
= (ax  dy)2
467
+ 4abedxy > O.
Therefore (O,O) s a
Exercises
10.2
= (3,4).
(e) Since 7 = 0.2, ,0,.= 0.03, and 72  4,0,.= 0.08, (O,O) is an unstable spiral point.
26. (a) The critical point is X = (1,1).
(b) Frorn the graph, X appears to be a center.
468
Exercises
10.3
Exercises 10.3
ddr = ~ (x ddx + y ddY) = ~(ax2  f]xy + xy2 + f]xy + ay2  xy2) = ~ ar2
t
r
t
t
r
r
Therefore r
ce'" and so r
+
= aro
< O.
dr
2. The differential equation dt = ar(5  r) is a logistic differential equation. [See Section 3.2, (4) and
e = t + e2.
and
1 + qe50t
g'(n + 1) = k(n
= k(n+
< O, r
1)  2kx, g'(O)
+
Oas t
= k(n+
1)
+
+00
> Oand
k  kln(x/k),
g'(k) = k
Theorem 10.2.
5. The only critical point is T
Theorem 10.2.
= To.
Since g'(T)
Now g(v)
=9
(k/m)v
= To is unstable by
< O. Therefore
> Oand so x
k(a
 f]
+ 2x),
g'(a)
,
2x)
+ k(f]
 x)(I  x)(l),
= a
+ be
and
= (a  bP)
+ (P
 e)( b),
g'(e) = a  be.
469
e is
Exercises 10.3
10. Since A
 k, g'(K2)
k/2
< O.
g'(X) = (2Y
2X)
2x  1 .
2y
= 2 and 6. = 2 so that 72
46. = 4
< O. Therefore
and
, (X) = (2XO 2Y2).
At X = (1, O), 7
= 4, 6. = 4, andso
72
are unable to classify this critical point any further. At X = (1, O), 6. = 4
a saddle point.
13. y' = 2xy  y
x
1/2 and y  x2
+ 2 = O implies
that (x,y)
(1/2, 7/4).
points (J2, O) and (J2, O). We next use the Jacobian matrix
g'(X) = (~~
1)
2x ~
= y + xy
= y( 1
y = O and 2x  y2
+ x).
= (1/2, 7/4),7
= 1,6.
= 1 and 6.
< O. Therefore
= 7/2 and so 72
46. =
(1, J2) and (1, J2). We next use the Jacobian matrix
g'(X) =
(2
2Y)
x  1
2,6. = 4, and so 72
+ 2 = (x
 l)(x  2)
points are (1,1), (1, 1), (2,2), and (2, 2). We next use the Jacobian
g'(X) =
(3
2x
470
2Y)
2y
Exercises 10.3
= (1,1),
= 5, 6 = 2, and so
(1, 1),6
T2 
46
= 17 > O.
= (2,2), 6 = 4 < O and so we have another saddle point. Finally, if X = (2, 2),
For X
6 = 4, and so
T2 
= 1,
= O, y = x or y = x. The case y = x leads to (4,4) and (1,1) but the case y = x
leads to x2  3x + 4 = O which has no real solutions. Therefore (4,4) and (1,1) are the only
16. From y2  x2
x
g'(X) = ( y
2y
2x
= (4,4),
= (1,1),
T
T
3)
< O. It follows
= 2xy = O, either x = Oor y = O. If x = O, y(l y2) = Oand so (O,O), (0,1), and (O,1)
are critical points. The case y = Oleads to x = O. We next use the Jacobian matrix
g'(X)
2y
( 1 + y
2X)
1+ x  3y2
= (0,1), T =
4, 6
= 4 and so T2
46
= 1 and
= O. We can
point but we are unable to classify this critical point further in ths borderline case. For X = (O,1),
6 = 4
18. We found that (O,O), (0,1), (O,1), (1, O) and (1, O) were the critical points in Exercise 15,
Section 10.1. The Jacobian is
g'(X) = (
For X
= (O,O), T = 4, 6
1  3x2  3y2
2xy
= 3 and so T2  46 = 4
= 8, 6 = 12, and
= (1, O) or
T2
6XY)
3 _ x2 _ 9y2 .
19. We found the critical points (0,0), (10,0), (0,16) and (4,12) in Exercise 11, Section 10.1. Since the
Jacobian is
g'(X) =
10  2x  Iy
2
y
471
_IX)
2
16  2y  x
Exercises
10.3
Conclusion
unstable node
(O, O)
(10,0)
26
160
4
(0,16)
14
60
32
(4, 12)
16
24
36
saddle point
saddle point
stable node
160
20. We found the sole critical point (10,10) in Exercise 12, Section 10.1. The Jacobian is
gl(X) = (
g'((10, 10)) has trace
= 46/15,6.
22
= 2/15, and
15
 (y + 5)2
72 
46.
yl = (cos e  ~) sin e.
is
gl(X) =
( cos 2e  ~ cos e
~)
and so at (O,0),7 = and 6. = 1/2. Therefore (O,O) is a saddle point. Por X = (11"/3, 0),7 =
and 6. = 3/4. It is not possible to classify either critical point in this borderline case.
Xl = y,
If (x, y) is a critical point, y =
+ (~_ 3y2)y
yl = x
and so x
+ x)
 x2 = x(l
_ x2.
= O. Therefore (O,O) and (1, O) are
the only two critical points. We next use the Jacobian matrix
gI (X)
(0
1  2x
1)
"2  9y
72 
46.
is an unstable spiral point. Por X = (1, 0),7 = 1/2, 6. = 1 and so (1, O) is a saddle point.
23. The corresponding plane autonomous system is
and the only critical point is (O,O). Since the Jacobian matrix is
gl(X) = ex
+03x2y x3
472
J'
Exercises
10.3
= 1 and 6. = O, and we are unable to classify the critical point in this borderline case.
4x
 2y
1 + x2
and the only critical point is (O, O). Since the Jacobian matrix is
I
x = y,
(4
10_x2
_:),
(1 + x2)2
12, and so (O,O) is a stable spiral point.
g'(X)
y = 
critical points.
g (X) =
to classify these three critical points. Por X
this critical point.
Por
(flTt, O),
1)
1 + 3tx2
= O and 6. = 2
saddle points.
26. In Exercise 6, Section 10.1, we showed that (0,0), (l/t,O),
Since Dxxlxl
21xl, the
6.
= (O, O),
Jacobian matrix is
g (X) =
Por X
and (1/10,0)
(O2Elxl _ 1 O1) .
= O, 6. = 1 and we are unable to classify this critical point. Por (l/E, O),
= O,
x'
y,
1 + 0:2x2
y =
= (
((3 + o:y2)~0:2x2  1)
(1 + 0:2x2)2
Por X = (O, O),
28. From x'
y
= 1/(3.
(Xx
= O and 6. = (3. Since (3 < O, we may conclude that (O, O) is a saddle point.
+ xy
The case y
= x(
o: +
= o: implies that
1  (30:  x2
= O or x2 = 1 
has no real solutions. It follows that (0,1/(3) is the unique critical point. Since the Jacobian matrix
is
g'(X)
(o: +
2x
473
X)
(3'
Exercises 10.3
T
= a  (3 +
1 a(3
73 = (3 + (3
< O and 6.
= a(3  1
point.
29. (a) The graphs of x+yx3
= Oand _x_y+y2
g'(X) =
( 1
3x2
1
1)
1+ 2y
T2 
= O are shown
(b) For Xj , 6.
= y,
X'
3
and so the only critical point is (O,O). Since the Jacobian matrix is
I
g (X) =
T
= E, 6. = 1, and so
T2 
46. = E2
(O1
1 ),
E(l _ y2)
(b) When
2x3
may be solved by separating variables. It follows that y2 + x4 = C. If X(O) =
dx
X
y
(xQ, O) where XQ > O, then C = x~ so that y2 = x~  x4. Therefore if XQ < X < XQ, y2 > O and
dy
y'

so there are two values of y corresponding to each value of x. Therefore the solution X(t) with
X(O) = (xQ, O) is periodic and so (O,O) is a center.
32.
~ i
dX
x'
~~may
y
(x1)(x22x2)
474
~ ~
 =   x2
2
3
+ C and
Exercises 10.3
But (x  1)(x2  2x  2)
> Ofor 1 J3 < x < 1 and so each x in this interval has 2 corresponding
2xy
O or y
= O.
If x
= O, then
from 1  x2
+ y2 = O, y2 = 1
g'(X) = (
and so
T =
O and 6.
=4
at either X
2y
2x
2X)
2y
1  x2 + y2
d
or 2xy.JI.. = 1  x2 + y2. Letting
2xy
dx
dJ.L 1
1
J.L=  , it follows that  =  1 and so J.L=    x + 2c.
dx
x2
X
x
y2
1
Therefore  =  x+2c
which can be put in the form
x
x
dy
dx
y'
x'
y2
(x  c)2
+ y2
= c2
l.
The solution curves are shown and so both (1, O) and (1, O)
2
are centers.
34. (a)
dy
dx
y'
x'
(b) Let w
x  y2
X
dy
=    y and so y
y
dx

+ y = xy
1
= yln = y2. It follows that dw + 2w =  2x, a linear first order differential equation
dx
whose solution is
y 2 = w = ce 2x.
(1 x ) .
+ "2 
="2 1 
x,
dy
dx
y'
x'
y2
y
and so
x4
x2
Cl
v5 and so
. 475
or y
x4
"2

+ Cl
Since x(O)
O and
Exercises 10.3
2 (x 2  1) 2 + 2v5  1
Y ="2x  x + Vo =
2
.
2
JI  2v5
= 1
) for real
solutions to existo If x5
< x < xo, then (x2  1)2 + 2V6  1 > Oand so there are two corresponding
1v2.
y' = X2
 x
+1
g (X)
to attempt to classify ((1 ~)/2,O)
For X
(O
2X _
1)
1 O
((1  vT=4E)/2,
1 vT=4E
2
= O and
6. = =f~.
O), 6. 2: O and we are not able to classify this critical point using linearization.
,
a
f] 3 R
y =  x   x   y
L
L
L
= y,
= q and y = q'. If X = (x, y) is a critica! point, y = Oand ax  f]x3 = x(a + {3x2) = O.
> O, a + f]x2 = O has no real solutions and so (O,O) is the only critica! point. Since
where x
If f]
= ( a _O3f]x2
g' (X)
= R/ L < O and 6. = a/ L > O. Therefore (O,O) is a stab!e critica! point. If f] < O, (O,O) and
(X,O), where x2 = a/{3 are critica! points. At X(X,O),
T = R/L
< O and 6. = 2a/L < O.
= y, then dy/dt
= x3
dy
dx
476
Exercises
10.3
=
y dy
(x3
+ x)
dx
and
1 2
1 4
1 2
+ 1)2
and so
2
1 (2
y = "2 x
Note that y
_~(x2
= O when x
)2
1 (2
)2
+ 1 +"2 xQ + 1
x6
+ 2x6 + 1 
x4
2x2  1
so there are two corresponding values of y for each x between XQ and XQ.The solution X
that satisfies X(O)
39. (a) Letting x
= Xl
Xl
= Y and
yl =
sin 5n /6 = 1/2 we see that (n /6, O) and (5n /6, O) are critical points of the system.
O 1)
(  cosx
O
and so
Al = gI = ((n/6,0))
(O
= V3/2
01) and
A2
= gI =
((5n/6,0))
(OV3/2 O1) .
> O and the trace of Al is O, no conclusion can be drawn regarding the critical
point (n/6,O). Since det Aj, < O, we see that (5n/6,0) is a saddle point.
Since det Al
(e) From the system in part (a) we obtain the firstorder differential equation
dy
dx
1/2  sinx
y
J
and
y dy =
J (~
sin x ) dx
1
1
_y2 = x
2
2
+ cos x + C
y2 = X
+ 2 cos x + C2.
or
477
Exercises
10.3
For Xo near
7r /6,
= (xo, O) then
if X(O)
C2
Thus, there are two values of y for each x in a sufficiently small interval around
(7r /6,
40.
7r /6.
Therefore
O) is a center.
x(x3  2y3) and y' = y(2x3  y3) we see that (O, O) is a critical
+ (2x4u3
x4u(1
+ u3)
+ x5(2u3
dx
2u3
 dx
x
1 dx
x
+ xdu) = O
 x4)(udx
 1) du

=O
+ u (3u + 1) du
+ (1  1 + 2u  1 )
u +1 u
u2  u + 1
=O
du = O.
Integrating gives
In Ixl + In lu + 11ln
lul + In lu2 
u + 11=
or
x (u: 1) (u + 1)
(y ; x) (~~ _ ~ + 1) =
= C2
2 _ U
(xy
+ x2)(y2
 xy
xy3
x3
478
c2
+ x2)
= c2x2y
+ x4
= c2x2y
+ y2 = 3c3xy.
Cl
Exercises
10.4
(c) We see from the graph that (O,O) is unstable. It is not possible
to classify the critical point as a node, saddle, center, or spiral
point.
Exercises 10.4
= e(o) = !!: and y(O) = e'(O)
2g
7f
cos '3
+ e = T + e and
so e
= Wo  T'
y2
Smce
y2
wo.
2g
T cosx
+ e,
Wo
Therefore
2t (cos x  ~
+ 2lg w5)
.
1
1 2
1
1 2
and the xmtercepts occur where cos x =    Wo and so    Wo must be greater than 1 for
2
2g
2. (a) Since y
2g
= T cosx + e, x(O)
~
l (cose  coseo).
When e
= e(O)
Iwol <
= eo and y(O)
2g
1Ff.
2g
= e'(O) = O, e = Tcoseo
and so y
~
= eo, y2 = (cos(
ea)  cos(eo)) = O. Therefore y
1
= 
&
= O
when e = eo.
(b) Since y = de and e is decreasing between the time when e = eo, t = O, and e = eo, that is,
t
=T,
dt
de = _
Therefore dt = de
dt
{Jg
T = _ (T
V 2g
fi9 J cos e VT
cos eo .
1
and so
2g ...jcos e  cos eo
r:
Jo=oo
1
de =
...jcos e  cos eo
(T rOo
V 2g
f'(x)
y' = g 1 + [f'(x)J2
479
 my
de
.
Exercises
10.4
and .
If X
~ (_
f'(X)
_ P_ ) __
(1 + [f'(x)J2)f"(x)
 f'(x)2f'(X)f"(X)
DX
9 1 + [f'(X)J2
my 9
(1 + [f'(x)]2)2
therefore
I
g (Xj ) =
4. When {3
(O
g
f"()
1)
(3.
m
Xl
~)
= J9f"(X1)i.
=O
has solution
x(t)
and period 21T/ J gf"(xl).
5. (a) If f(x)
x2
=
""2'
j'(x)
= x(O)
cos J9f"(X1)
= x and so
dy
dx
y'
x'
X 1
1 + x2 y.
==g
+ x2) + c.
Now
V5  gln
(1 ++ x~)Xo
1
= vD2 
2)
1+ x
1 + Xo
9 In
2
+ X5)  1J1/2, there are two values of y for a given value of x and so
x2
""2'
the maximum height occurs at the largest value of x on the cycle. From (a),
= [ev5/9(1
6. (a) If f(x)
= coshx,
X2
max
= _[e
2
(1
vO/9
+ x5)  1J.
480
Exercises 10.4
.
=  2g + e.
But x(O)
cosh x
Xo and
2
2g
Therefore e = Vo  and so
cosh Xo
2
2g
29
2
Y  +vo
cosh x
cosh Xo
Now
2g
cosh x
_
2g
cosh Xo
 
2g cosh Xo
 2g cosh Xo
2
+ Vo > O if and only if cos11X < '___,__
V6
has two
= cosh x, the maximum height occurs at the largest value of x on the cycle. Frorn (a),
= a where cosh a =
2g cosh Xo
2
2g  Vo cosh Xo
Zmax
7. If Xm
Therefore
2g cosh zr,
2
2g  Vo cosh xo
F(xn).
Letting X = Xl,
G(y) = ~
= F(xm)G(alb)
F(Xl)
F(Xl)
eo
< G(alb).
Xn
there is no corresponding
value of y. Therefore the maximum number of predators is Xn and Xn occurs when y = alb.
9. (a) In the LotkaVolterra Model the average number of predators is die and the average number
of prey is alb. But
Xl = ax + bxy 
EIX =
+ EI/b).
(b) The average number of predators di e  E21e has decreased while the average number of prey
alb + EI/b has increased. The fishery science model is consistent with Volterra's principie.
481
Exercises 10.4
+ 0.02y) = O
x( 0.1
y(0.2  0.025x) = O
in the first quadrant we obtain the critical
20
40
60
100
80
= 4.
was
Since x(O)
in M athematica.
to
approximate the solution of x(t) = 7 for t near 40. From this we see that the period is more
closely approximated by t = 44.65.
11. Solving
x(20  0.4x  0.3y)
=O
=O
we see that critical points are (O, O), (0,100), (50, O), and (20,40). The Jacobian matrix is
'(
X)=
0.024x
)
0.06(10  0.2y  0.3x)
and so
Al = g, ((O, O)) = (1.6
A3
= g, ((50, O)) =
(1.6
O
O)
0.6
1.2)
0.3
, ((0,100)) = (0.8
1.8
A2
=g
A4
= g'((20, 40))
0.64
= (
0.72
O)
0.6
0.48)
0.24
= 6.1 = 0.96 > O, T = 2.2 > O, and T[ 46.1 = 1 > O, we see that (O, O) is an unstable
node. Since det(A2) = 6.2 = 0.48 > O, T = 1.4 < O, and Ti 46.2 = 0.04 > O, we see that (0,100)
is a stable node. Since det(A3) = 6.3 = 0.48 > O, T = 1.9 < O, and Tl 46.3 = 1.69> O, we see
that (50, O) is a stable node. Since det(A4) = 0.192 < Owe see that (20,40) is a saddle point.
Since det(Al)
12. 6.
= rlr2,
= (rlr2)2.
r2, (0,0)
is an unstable node.
13. For X
T2
= (Kl, O), T = rl +rz (1  ~~ Q2l) and 6. = rlr2 (1 ~~ Q2l). Ifwe let e = 1 ~~ Q21,
 46. = (cr2
+ rl)2
stable node. If Kl
482
Exercises
14.
10.4
(x, f)) is a stable no de if and only if K1 > K2 and K2 > K1. [See Figure 10.38(a) in the text.]
012
021
From Problem 12, (0.0) is an unstable node and from Problem 13, since K1
< K2 , (K1, O) is a
021
of 1 and 2 interchanged. Therefore (O, O), (K1, O), and (O, K2) are unstable.
K1
15 . 012
1. T here fore
/\ = (1 
t...l
T1T2
012021 )"xy K1K2
a saddle point.
16. (a) The corresponding plane autonomous system is
X'
= y,
I
g.
y = smxy
(3
ml
and so critical points must satisfy both y = O and sinx = O. Therefore (mf,O)
points.
are critical
_1(3)
( _E.:osx
Z
ml
has trace
.6. =
2
T
E. > O at
9
(32
4.6. = m2Z2  4 T
(32  4gZm2
m2Z2
(32  4gZm2
< O or
(3 < 2m.gl.
= y,
(3
y = yly
m
I xk
and so a critical point must satisfy both y = O and x = O. Therefore (O, O) is the unique critical
point.
(b) The Jacobian rnatrix is
( _ k _
m
and so
/2IYI)
= O and .6. = k > O. Therefore (O, O) is a center, stable spiral point, or an unstable
m
spiral point. Physical considerations suggest that (O,O) must be asymptotically stable and so
(O,O) must be a stable spiral point.
483
Exercses 10.4
18. (a) The magnitude of the frictional force between the bead and the wire is .t(mgcosB)
for sorne
.tmg
/1
"
mx
f'(x)
.t
x + mg 1 + [f'(x)]2
(3 ,
1 + [f'(x)J2 
= mg
and so
"
x
.tJ'(x)
= 9 1 + [f'(x)J2
(3,
 mx .
= u:
m~)
g'(X) = (9f~(XI)
m~)'
1 + .t2
(3
Therefore r = 
occurs. When
< O and
,0.. =
9 f" (x )
~ . When
1+.t
f"(XI)
(32
,
m2
4g
f"(xI)
< O,
1 + .t2
19.
ddy
y:
f(x)
Y
(O,O) were a stable spiral point there would exist an x with more than two corresponding values of
484
Exercises
20. (a) x'
10.4
= O, then, from
exy + K y(K  y) = O,
y
= O or K.
ex = K
(K  fj).
g'(X) = (
and so at X
a + by
ey
> b' 6. < O in this case. Therefore (O,O) and (O,K) are each saddle
y
= K
Y < Oand
6. = bcifj
and ci: = ~ (K 
g'((x, fj))
T
1)' Since
and so
bx
T
>
= (
,
ey 
bX)
T,
K
O. Therefore (x, y) is a stable critical point and so it is
using
ci= ~(Ky)=
Therefore
72 
46.
~Y[(~+4b)Y4bK].
4bK2
K + 4b = T + 4bK
Note that
4bK2
T+4bK
1< K.
4bK .K::::K
T+4bK
Therefore
72 
485
46.
Exercises
10.4
=a
x'
implies that x
o or y
= _1_.
1~
Y x  x
When a
a1
=x
y 
1) = o
> O, Y = _1_
e7
It follows that
x=
and if (3(a  1)
= _a_[(a
1)(3 1]
a1
a1
1, x > O. Therefore (x,?)) is the unique critical point in the first quadrant. The
>
a ((3  _1_)
Jacobian matrix is
:x
a y ~ 1 1
(1 y) 2 )
x
(
y
 1
l+y
(1+y)2
and for X = (x, y), the Jacobian can be written in the form
g/(X) =
(a ~ 1)2 x
).
(a  1)2 _ 1
a2
It follows that
2
7
and so
+ 1).
= (6
= _ [( a :21)
Therefore 72  46
x + 1] < O,
= (6 + 1f  46
(a :21) 2 X
(x, y) is a
stable node.
22. Letting y = Xl we obtain the plane autonomous system
Xl
=Y
yl = 8x
Solving x5
+ 8x =
6x3
+ 6x3
x5.
x(x2  4)(x2  2)
O we see that
critical points are (O, O), (O,  2), (0,2), (O, y2), and (O, y2).
The Jacobian matrix is
I
g (X)
8 + 18x2 _ 5x4
1)
and we see that detg'(Xj) = 5x1_18x2 +8 and the trace of g/{X) is O. Since det(g'((y2,
8
O))) =
< O, (y2, O) are saddle points. For the other critical points the determinant is positive and
linearization discloses no information. The graph of the phase plane suggests that (O, O) and (2, O)
are centers.
486
Chapter
10 Review Exercises
1. True
2. llue
3. a center or a saddle point
4. complex with negative real parts
5. False; there are initial conditions for which lim X(t)
t=ecc:
(O,O).
6. llue
7. False; this is a borderline case. See Figure 10.25 in the texto
8. False; see Figure 10.29 in the texto
9. The system is linear and we identify 6. = o: and
when 6.
> O and
= Owe see that for o: = 1 critical points will be centers and solutions will be
= x  2y
yl = X
+ y,
Xl
dt
dt
di
r2
di
di
r2
1
Using separation of variables it follows that r = ij3TtCl and
r
= 1 and
e = O,. It
3t + el
follows that
el

00,
+
Since X(O)
= (1, O),
= 1, e2 = O, and so
r=
As t
e = t + e2.
ij3t + 1 '
e = t.
= Xa lies on the line y = 2x, then X(t) approaches (O,O) along this lineo For all other
initial conditions, X(t) approaches (O,O) from the direction determined by the line y = X.
487
Chapter
10 Review Exercises
(b) If X(O) = Xo les on the lne y = x, then X(t) approaches (O,O) along this lineo.For all other
initial conditions, X(t) becomes unbounded and y
13. (a)
= O, 6. = 11
(b)
= 2, 6. = 1, 72
= 2x serves as an asymptote.
46.
= O, then, from y( 4  2x  y)
=O
If 1 + y  3x = O,
then y(5  5x) = O. Therefore (1/3, O) and (1,2) are the remaining critical points. We will use the
Jacobian matrix
g' (X) =
(1 +
y  6x
2y
x
)
4  2x  2y
(O, O)
15. From x
(0,4)
4
20
(t, O)
la
3
(1,2)
saddle point
saddle point
10
15
dx
= rsine
dt
de
dt
5
dy = reose de
dt
dt
Then r'
Conclusion
unstable node
dr
cose
dt
+
+ dr
sine.
dt
= or , e' = 1 gives
dx
= r sine
dt
+ ar cose
. e.
dy = T cos e + Cl.r sm
dt
We see that r
r
r' = Oand r = Cl. In this case (O,O) is a center, which is stable. Therefore, (O,O) is a stable critical
. point for the system when CI. ::; O.
16. The corresponding plane autonomous system is x' = y, y' = t(1  x2)  x and so the Jaeobian at
the critical point (O,O) is
g/((O,O))
= (_~ ~).
488
10 Review Exercises
Chapter
= J.L,!:::. = 1 and
72 
dy
dx = ,x =
x(O)
= Xo
+1
_2xVy2
y
 4x  1),
= O. It
_x2
+ c.
But
= (1 + x6  x2)2 
l.
Note that 1 + X5  x2
[Y2+l =
= xo.
has two corresponding values of y and so the solution X(t) with X(O)
x' = y,
+ x )3 + 9
g(X)=
Por X = (O,O), 7 = _f}_
O
~(s+x)2
7 
(32
12k 2
1
 s = ((3
m2
m
m2
41:::. = 
 12kms
).
= mg.
= g/w2l.
o.
= O) of
COSX
But if w2
g'(X) = (
and so
= (3/ml
< gil,
 gil)
O
w2 cos 2x 
g/w2l
1:)
T cos x :%J
asymptotically stable and so after a small displacement, the pendulum will return to
e' = o.
489
(O,O) is
e = O,
Chapter
(b) If w2
(x,
10 Review Exercises
that satisfy
saddle point. If X =
6. =
cos x
(x,

O), T
7r
w2 cos 2x
3:2  (2 :':2  1) =
w2
w2
3:2 >
= ea, e'(O) =
o and
expect the pendulum to reach one of these two stable equilibrium positions.
490
O.
ea is small we
11
1.
2.
r2 xx2dx
= ~x412
r1 x3(x2
+ l)dx =
12
J1
42
102 eX(xeX
4.
6.
571"
~x611 +~x411 =
6
1
4
1
3.
 eX)dx
5. rr/2 xcos2xdx
= ~ (21 cos2x
+ xSin2x)
1:=0
 x)
171"/2
71"/2
=O
h~
7I"~
7. For m =1 n
rr/2
Jo
= 1 1071"/2 [cos2(n
2 o
1
4 nm)
= (
171"/2
sin2(nm)x
(.
ti
+ m + 1 )sin2(n+m+1)x
= O.
For m = n
Jorr/2 sin2(2n+1)x
dx =
r= (12  21
Jo
1
=x
2
11[/2
o
1T
=
491
) dx
1
) sin2(2n+
4 n+1
(2
l)x
11[/2
o
171"/2
o
Exercises 11. 1
so that
II sin(2n
8. For m =1
+ l)xll =
2"V1f.
ti
rr/2
Jo
=
1
(
4nm
17r/2
)sin2(nm)x
1
(
4n+m+l
17r/2
)sin2(n+m+l)x.
= O.
For m = ti
fo7r/2
1~2
= x
2 o
1~2
(2
) sin 2(2n + l)x
4 n+l
o
so that
II cos(2n +
l)xll =
2"V1f .
9. For m =1 n
7r
lao
sin nx sinmxdx
la
2nm
17r 
)sin(nm)x
( 1
2n+m
17r
)sin2(n+m)x
= O.
For m
=n
7r
lco
sin2nxdx
1
] dx
= lo7r [1
  cos2nx
o 2 2
so that
II sin nxll
492
1
= x
2
;_.
7r o
1
sin2nx
4n
17r=7r
o
Exercises
10. For m
11. 1
=1p
n?f
mit
P
sin x
dx
20
p
2(nm)Tt
. (nm)7r
p
S111
IP 
p
. (n+m)7r
sm
x
2(n+m)7r
p
IP
O
= O.
For m =
ti
(P
Jo
[1'2  1
2 n7r
(P
sin ;xdx = Jo
2n7r]
'2cospx
dx
= '2x
I0P
p
2n7r
4nTtsinpx
IP0= p'2
so that
loO
lloP ( cos (n 
n7r
mt:
cos  x cos  x dx = p
20
m)7r
x + cos
p
. (nm)7r
IP
sin
x
2(nm)7r
p
o
(n
+ m)7r x )
p
dx
p
. (n+m)7r
sin
x
2(n+m)7r
p
= O.
For m = n
p
loo
n7r
cos2 xdx=
p
loP (1+cosx
1 2n7r)
O
dx=x
Also
p
loO
nst
1 . cos x
p
4n7r
IP = O
p
n7r
dx =  sin x
n7r
p
IP +sinx
p
2n7r
p
IP =.p
O
and
so that
12. For m
=1
j
j
p
P
p
P
n7r
cosx
m7r
cosxdx
n7r
sin x
sin x
n7r
cosx
mt:
cosxdx
p
= 2 lcP
n7r
sin x
m7r
sin x
tnat
= 2 JP
dx
=O
dx = O.
Also
nm
j Pp sin x
P
tnn
cos x
P
1
dx = 2
jP ( sin (n P
493
m)7r
P
x + sin
(71+ m)7r x )
P
dx = O,
IP
O
Exercises 11. 1
P
P
mr
1 . cos x
P
mr
P
n7r
sin x
P
dx =
l sm x
P
and
n7r
cos x dx
P
dx =
P
P
dx =
n7r
Pp sin 2 x
P
dx = P
P
P
and
IP = O,
mr
sin x
p
mr
p
ru:
 p cos x
IP
mr
= O,
P
2n7r
p
2n7r
sin x
dx =  cos x
p
4n7r
p
Por m = n
x
Pp cos2 n7r
P
p
(1 + 1
(1 1
cos 2n7r)
x
p
dx = p,
2n7r)
   cos x
dx = p,
IP
P
= O.
12dx= 2p
P
so that
13. Since
CXlex.2
l. 2xdx = _ex
~
OC ex 2 1
CXl
210
e x21CXl=
O
00
(4x2  2) dx = 2
=2
OC
CXlx ( 2xex2)
dx  2
O
'
CXl
2
oc ex dx
21C CXl 2)
( xex CXl + CXlex dx
 2 CXl ex 2 dx
oc
and
2) dx = 4
OC
(
2)
oc x2 2xex
= 4 ( _x2ex 21CXl
oc +
494
dx  4
OC
CXl xe"?2dx
CXl
CXl xex 2)dx
 4
CXl
CXl xe?" 2dx
Exercises 11. 1
14. Since
Jo
foOO ex
1 Gx2
 2x
x) dx
+ 1) dx
= (x
OO
roo
O  Jo
 l)eX
= (2X 1
= 1 + (2
= O,
I~+ fooo
~x2) ex
 x)eX
exdx
I~+ fooo
exdx
eX(x
 2)dx
= O,
and
= e=
(1"2X "2x5
3
+ 3x  1) 100
o + Joroo e x
1 + ex Gx2  5x + 3) [ +
=1 3
+ eX(3x
 5) [
3
fooo
fooo
23x 2 + 5x
3) d x
eX(5  3x) dx
exdx
= O,
15. By orthogonality
n = 1,2,3, ....
(o:x
; that is,
Ji <Pn(x)dx = O for
+ (J)<Pn(X)
dx
= o:
x<Pn(x) dx
+ (J l; t
{b
<Pn(X) dx
(b
= o: Ja <Pl(X)<Pn(x) dx + (J Ja <PO(x)<Pn(x) dx
=0:0+{J0=0
for n = 2, 3, 4, ....
m we have
= b <p~(x)dx + 2l
= lI<Pm(x)1I2
=1=
<Pm(X)<Pn(x)dx
+ 11 <Pn(x)
112 .
495
+ b
<p;(x) dx
Exercises 11. 1
18. Setting
(2
12
0=
(2
+ CX3 + C2X )
dx =
16
64
:3 + SC2
and
0= [:
we obtain
= O and
CI
13(x)h(x)
C2
dx
(X3
[22
+ CX4 + C2X5) dx
= 654CI
= 5/12.
[7r,
7r],
= 1 is orthogonal to every member of {sin nx}. Thus {sin nx} is not complete.
and f(x)
20. (h
(b
+ h, 13) = Ja
+ h(x)l13(x)
[h(x)
(b
= Ja h(x)13(x)
dx
27r / 27r
27r / ( 4/ L)
dx
(b
+ Ja
h(x)13(x)
dx
l.
= 17r L.
+ sin 2x is 27r.
period of sin 2x + cos 4x is 27r / 2 = tt .
period of sin 3x + cos 4x is 27r since the
27r /
= 2p/n.
is 27r/(n7r/p)
Exercises 11.2
1.
ao =
an
I.
i" f (x)
7r J"Ir
1
bn = 
"Ir
j"lr
"Ir
7r
f(x)
100
7r n=l
7r
I.
7r
1
[1(ltl
n7r
tit:
sinnx
sinnxdx
=O
(0 1 dx
J "Ir
+ I.
7r
r 2dx = 1
Jo
bn
1 lo"lr
7r
l(l)n
11"
=
7r
= 1
7r
n7r
f(x)sinxdx
=  +
= 1 J"Ir cosnxdx
7r
2. ao = I. j11" f (x) dx
an
I.
i" 1dx
7r Jo
dx =
= 1 j"lr f(x)cosxdxn7r
7r
"Ir
= 1 j11" f(x)sinnxdx
7r
11"
27r /3
7r
11"
= 1 JO sinnxdx+
7r
rr
7r
1 lo"lr 2sinnxdx
7r
496
3
= [1(ltl
n7r
Exercises
1 3 00 1  (_1)n
f(x) =  +  L
2 7r n=l
n
3. ao = JI f(x) dx
1
SlD nx
= JO 1 dx +
1
Jo
x dx
=~
2
an
1
= J I f(x) cosn7rX dx = JO coenxx dx + 101 x cosrcs x dx = 22"[(lt
1
1
O
n 7r
bn
3
f(x)=4+L
00
[(1)n1
n=1
4. ao = I f (x) dx =
an
b, =
1
J
J
1x
f(x)cosn7rxdx
Ion
I
1
5. ao = ~j7r
[(
f(x)dx
tt
i7r
x2 sin nx dx
f(x) = 7r2
(_l)n+l
cos n7rX +
(_1)n+l
[2(l)n
n2
n=1
00
]
sin n7rX
n7r
r x2dx
1
= ~7r2
i7r
7r
7r
non
tt
cosnx
(x
1 sinnx
2
x2cosnxdx=
= 1 (  x2 cos nx
+L
lJ
tt
7r
7r
7r Jo
1 j7r f(x)cosnxdx=7r 7r
bn = 1
dx = ~
2
=~
n7r
n7r
_l)n  1
2
1
cosn7rxsinn7rx
7r
an=
1
3
00
f(x) = 4 + L
n=1
17r+ 2lo7r
17r  2lo7r
7r
x cos nx dx )
+ (7r'_(_l)n+l + 2[(l)n1J)]
n
n37r
nOn
sinnx
7r
non
7r
7r
1 JO 7r2sinnxdx+
1 lo7r (7r2x2)sinnxdx
7r
Ti
7r
497
xsinnxdx
2( _l)n
=2
2 [(1t
= tt (1)n+ 1+ 3
6. ao = ~
an=
lJ
22
7r
Jo
11.2
tt
 1J
Exercises 11.2
= ~[(lt
11"
J; [2n2 (_I)n+l
511"2
f(x) = (3 +
11"
11"
11"
I.
an = 1
11"
00
I,
j1l"
11"
= 211"
=O
= 1
11"
I,
'(
(3  2x) cos nx dx
=O
11"
4
n
(32x)sinnxdx=(lt
11"
= 3 +4 L
=6
11"
f (x) cos nx dx
n=l
_1)n

sin nx
r sin x dx = ~11"
11" Jo
11"
= 1 j1l" f(x) cosnx dx = 1111" sinx cosnx dx = 1 la1l" [sin(n + l)x + sin(1  n)x] dx
11"
11"
11"
1+ (lt
= 1I"1n
(
2)
al
11"
(3  2x)dx
= .:
11"
11"
(lt]
00
f (x)
sin nx
n 11"
00
11"
bn=11"
(In])
+ 2[1 n311"
+ i[I
2
L (
_l)n+1 sin nx
= 11" +
= 1
= ~(lt
8. ao = .!.j1l" f(x)dx
11" 11"
an
11"
11"
n=l
9.
+ (11"~(_I)n
r iCOSnXdX)
Jo
f (x) cos nx dx
11"
11"
11"
f(x)
an
cos nx
7. ao
111"
nOn
= 211"
Jor
for n
sin2xdx
211"
= 2,3,4, , , ,
=O
bn = 1
f (x) sin nx dx = 1 la1l" sin x sin nx dx
11" 11"
11" o
1 {11"
= 211" Jo [cos(1  n)x  cos(1 + n)x] dx = O for n = 2,3,4, , , ,
b: = 21
(1  cos 2x) dx = ~
11" k
2
f (x)
1
=
11"
ool+(_I)n
n=211"1  n
+  sin x + L
2) cos nx
498
Exercses
10.
11.2
2 j1r/2
2 Jc1r/2
?
f (x) dx = cos x dx = ::
7f 1r/2
7f o
tt
ao = an
= 2 j1r
tt
/2
1r/2
= 2J1r
/2
= 2Jc1r
/2
7f
1
cosx cos 2nx dx = 7f
!n
1r/2
[cos(2n  l)x
+ cos(2n + l)x] dx
2(_1)n+1
7f(4n2  1)
2 j1r /2
bn = f(x) sin 2nx dx
7f 1r/2
7f
1 Jc1r/2
cos x sin 2nx dx = [sin(2n  l)x
7f O
+ sin(2n + l)x] dx
4n
7f(4n2  1)
1 L [2( _1)n+1
4n
00
f (x) =  +
(2
) cos 2nx
7f n=l 7f 4n  1
11.
0,0
= 21j2
2
an=21 j2
2
f (x) dx = ~ (JO  2 dx
2 1
f(x)
12.
ao
= 
;:2
t.
=  t.
=~
an=21
bn
1
2
+L
00
n=l
Jo
(10
dX) =  ~
2
1
2
1
2
ten
f(x)cosxdx=2
1
2
2
n7f
f(x)sinxdx=
2
1
2
ten
la
) =sin1
nr:
cosxdx
2
n7f
sinxdx
2
nat
= 3
ru:
n7f
2
( 1cos n7f)
2
n7f) . nr: ]
1  cos smx
2
2
2 cosxdx
n7f
2
n7f
xcosxdx+
2
O
l
O
+ 3
] dx+
la
\ dX) = ~
(!nI
(la
] dx+
1
X dx
 1
(JO [ 2cosxn7f
1 . n7f
nr:
n7f
sm  cos x
2
2
f (x) dx = ~
11
) sin 2nx
1
2
nr:
f(x)cosxdx=2
nr:
bn= 1 j2 f(x)sinxdx=2 2
2
+ 7f(24n
) =2"2
2 ( cos1
n7f
n tt
2
n7f
xsinxdx+
2
2
1
nr:
sinxdx
2
2 ( sin n7f + (
n7f lt+l )
= n27f2
2
2
f(x) = 3
8
+ '"
2 ( cosnr:
s: n27f2
2
00
n=l
n7f
an=51 j5 f(x)cosxdx=5
5
n7f
1 ) cosx+
2
1
5
(JO
5
n7f
cosxdx+
5
nr:
 2 (.smn7f + (1)
n27f2
2
2
la (l+x)cosxdxn7f)
5
499
n+l).
smxn7f ]
2
5
=2"2[(1)n1]
n 7f
Exercises
11.2
mf
bn= 1 j5 f(x)sinxdx=5 5
5
(JO
1
5
5
9
Loo' [ 5 [( 1)n lcosx+]
n1f
=+
4
n21f2
5
fx()
t.
2 2
n1f
f(x)cosxdx=
n1f
bn= 1 j2 f(x)sinxdx=2 2
(JO
2
(JO
2
la
n1f
(2+x)cosxdx+
2
O
la
n1f
(2+x)sinxdx+
2
O
n1f
2cosxdx
n1f
2sinxdx
2 [ 1  ( 1 )n] cos x
mf + 2 ( 1 )n+l.sm x
n1f ]
= 32 + Loo [ 2
n1f
2
n=l n21f2
bn =
16. ao =
an
1 j1l'
1 j1l'
f (x) sin nx dx = 
1f
11'
~ j1l' f (x) dx = ~
1f
1f
11'
bn =
f(x)
r(eX 
lo
1 j1l'
f (x) cos nx dx
1f 11'
=
1 j1l'
1f
f (x) sin nx dx
11'
e1l'1f1
21f
+L
00
n=1
eXsin nx dx =
11'
1) dx
= ~ (e 11'1f
1111'
(eX  1) cosnx dx
1 lo1l'
(eX  1) sin nx dx
[e1l'(_l)n_1
2) cosnx
1f 1 + n
')
2 2[1(lt]
n 1f
w
) =(lt+1
2
n1f
1f  1)
[e1l'(_l)n  1]
(
2)
1f 1 + n
1f O
1f O
(_l)nn(e1I'  e1l')
(
'))
1f 1 + n
==
f ()x
1f
5
=(lt+1
n1f
5
O
5 ( 1)n+l.smxn1f ]
n1f
5
n=l
an= 1
la (l+x)cosxdxmf)
5
n1f
sinxdx+
. 5
1 (ne1l'(_l)n+l
1
2
+n
=
1f
n
+ ( 2
1+ n
[
e1l'(_l)n+
n
(_l)n
+ 2
+  1)
1+ n
n
n
+ 11 +
(_l)n_1)
n
sin nx
17. The function in Problem 5 is discontinuous at x = 1f, so the corresponding Fourier series converges
500
Exercses 11.2
to 7r2/2 at x = Ti. That is,
7r2 7r2
= +
2
6
[2( _l)n
L
2
n= 1
n
00
cos n7r+
(7r
?[( l)n  1])
]
 (_l)n+l + 3
sin n7r
n
n tt
'
and
~2 = ~ (7r22_ ~2) = 1 + 212+ ;2 + ....
At x
so
7r2
1
1++
12 22
1
32
1
42
f (7r)
 = 7r+
2
(i
n=l n
7r
111
=1++.
4
357
1
2
tt
1 1
2
=+++_
tt
2 37r
and
or
=  +  + L.._.,
n=2
1 + (_l)n
nt:
cos 7r(1 n2)
2
2
3 . 57r
Ti
2
2
7r=1+++'"
2 3 35
2
57
Ti
1
1
1
4=2+~'3.5+57
501
2
5 . 77r
...
Exercises
11.2
= ao/2,
f(x)
L (nT
anCOS x
ao
2"
+
00
= Co
+L
einrrx/p
+ ibn)/2 we have
nT )
+ bnsinx
00
n=l
= (an
+ einrrx/p
2
an
+ bn
einrrx/p _ einrrx/p)
2i
 bn
ieinrrx/p _ ieinrrx/p)
2
n=l
00
= Co
+L
einrrx/p
+ einrrx/p
2
an
n=l
~
=co+L.._.
n=l
+ an + ib e inrrx/p)
2
00
= Co + L (Cneinrrx/p + c_nei(n)rrx/p)
n=l
00
eneinrrx/p.
n=oo
(b) Multiplying both sides of the expression in (a) by eimrrx/p and integrating we obtain
f(x)eimrrx/pdx
: ( f:
L
00
jP
Cn
n=oo
ei(nm)rrx/pdx
ei(nm)rrx/pdx
L Cn jP
ei(nm)rrx/pdx
p
ni'm
=
dx
p
L Cn jP
ni'm
cneinrrx/peimrrx/p)
n=oo
+ Cm jP
ei(mn)rrx/pdx
+ Cm jP
dx
p
P
L Cn jP
ni'm
P
ei(nm)rrx/pdx
+ 2pcm.
P
Recalling that
eiY=cosy+isiny
we have for nm
j p ei(nm)rrx/pdx
an integer and n
=
P
and
f. m
P
ei(nm)rrx/p
i(n  m)T
eiY=cosyisiny
\P
P
(ei(nm)rr _ ei(nm)rr)
i(n  m)T
=.(
p ) [cos(nm)T+isin(nm)Tcos(nm)T+isin(nm)T]
znmT
= Q.
502
Exercses
11.3
Thus
j P f(x)eim7rx/Pdx = 2pcm
. / "dx.
Cm = 1 jP f(x)e,m7rx
p
and
2p
= ex
P
+ 1)7r
2(in
+ 1)7r
=_
7r
7r
e(in+l)x
1
2(in
[e(in+l)7r
_ e(in+l)7r]
e(in+l)7r _ e(in+l)7r
= ;,
2(in
e7r(cos n7r
+ 1)7r
+ i sin n7r)
e7r
==~:..,:2(in + 1)7r
cosn7r
2(in + 1)7r
(e7r  e7r
= ';
(e7r  e7r)(_I)n
2(in
Thus
f(x) =
P'7r
'L
00
(lt e 2
n=oo
2(in
+ 1)7r
e
einx.
+ 1)7r
Exercises 11.3
1. Since f(x)
= sin(3x)
2. Since f( x)
= x
7. For
8.
= x + 5 = f(x),
9. Since f(x)
503
Exercises 11.3
10. Sincej'( x)
bn
=~
(1r 1. sin nx dx
~Jo
Thus
f(x)
00
n=l
rtt;
L
[1 
= 2. [1n~
(1 t] .
1dX
=1
n~
2
nt:
cos x dx =   sin 2 .
2
nt:
Thus
1
f(x) =  +
2
2
L
sinn=l ni:
nt:
00
nt:
cos x.
.:
xdx=~
n ~
Thus
~
f (x)
= +
2
00
1].
L 2[(1 t
n=l n ~
 1]cos nx.
Thus
00
n=l
2 lt+1.
= (
L (1t+
f (x) =
sin nx.
2
2
x dx = 
ao = 2
inO
an=2
x2
lo l x2cosn~xdx=2 ( sinn~x
n~
Thus
f(x)
= +
3
00
L 22(ltcosn~x.
n=l n ~
504
4
= 22"(lt
n ~
Exercises
16. Since f(x)
11.3
x2
= 2 ( cosn7rX
nt:
f 1 x2snmfxdx
)0
11
Thus
f(x) =
L (2(lt+1
n7r
tt
aO = 
tt
an
4
+ 33"[(
_l)n
00
n=l
= 2101r (7r2
7r
2 dx
X )
4 2
= 7r
3
= 2
x2)cosnxdx
(7r2
7r
4
= 2(lt+1.
n
Thus
f(x)
_7r2 +
3
4
L 2(lt+1
00
cosnxdx.
n=l n
= 2101r x3sinnxdx
7r
211"2
n
= _(_1)n+1
211"2
= (1)
ti
= 2
tt
( cosnx
x3
non
11r
+ 3101r x2cosnxdx
12 107r
xsinnxdx
n 11" O
2
+ 1
io
1r cosnxdx )
= 211"2(1
n'
)n+1 + 12 ( 1 )n .
n3
Thus
12 ( 1)n). sin nx.
f (x ) = ~L.. (211"2(
1)n+1 + "3
n=l
n
n
19. Since
= tt2101r
(x+
o
l)sinnxdx
2(7r + 1)
(_l)n+1
n7r
Thus
f(x) =
L
00
n=l
(2(11" + 1) (_l)n+l
n7r
505
+ 2)
n11"
.
smnx.
2
+.
nr:
Exercises 11.3
20. Since f(x) is an odd function, we expand in a sine series:
b., = 2la1 (x  1) sin mfX dx
= 2 [la1
]1 =  2
Ix
1
= 2 [ 22
sin mfX   cos mfX +  cos n7rX
n 7r
n7r
n7r
n7r
n7r
)
cos   1 .
2
Thus
f(x)
00
n=l
n7r
L sinn7rx.
=
an =
l
lao
l
x dx
1dx
= 3
n7r
nt:
cos xdx
2
4
= 22
nt:
Thus
4 (
nt:
)
nt:
f (x) = 43 + '"L00 cos   1 cos x.
n
27r2
2
2
n=l
= 1
7r
i7r
O
127r 7rsin x
n dx = 24 sin n7r + (
2 _l)n+l.
7r
2
n 7r
2
n
n
2
x sin x dx + .
Thus
f(x)
L (4sin
n7r + _(_l)n+
2
2
n
00
n=l n27r
1) sinx.n
2
ao
= 2
an
= 2 j.7r sinx
7r o
7r o
cosnxdx
la
7r(1:n2)[1+(ltl
al
forn=2,3,4,
...
io
= 1 7r sin2xdx = O.
7r O
Thus
f(x)
=~+
7r
f 2[17r(1+(_l)n]
n
n=2
2)
506
cosnx.
Exercises
24. Since (x) is an even function,
we expand
in a cosine series.
of Problem
of the integrals.]
ao =  2 la1f /2 cos x dx = 4
7r/2 o
7r
2
= /7r 2
an
Thus
tt 2
2
(x)=+L
7r
dx =
00
4(_l)n+1
n=l
7r 4n  1
(2
4(_l)n+1
(2
7r 4n  1
)cos2nx.
f1/2
25. ao = 2 Jo
1 dx = 1
1/2
an = 2
bn
=2
lnO
1 . sin n 7rx dx
00
n=l
L 2
00
J1/2
(x)
cos n7r)
+L
26. ao = 2 f1
1
1
1/2
bn = 2
= 2 ( 1 tin
n=l nr:
an=2
tin
1/2
1
(x) = 2
(x)
2
n7r
1 . cos n7rXdx =  sin 
lnO
1/2
1 dx = 1
lcosn7rxdx=sin
trn
tu:
ru:
= +
00
n=l
27.
+ (_1)n+1 )
sin n7rX
7r
4lc1f/2
2lc1f/2
an = 
7r O
bn =
4101f
7r
O
/2
2la1f
[cos(2n
/2
[sin(2n
+ l)x + cos(2n
+ l)x + sin(2n
507
 l)x] dx =
 l)x] dx =
4( _l)n
(
2)
7r 1  4n
(2
8n
7r 4n  1
11.3
10 in
Exercises 11.3
2
4( _l)n
00
f(x)
f (x)
8n
00
L
n=l7r
(2
) sin 2nx
4n  1
an= 2
tt
Jo
tt
lo'" sinxcosnxdx=
tt
lo'" [sin(n+1)xsin(n1)x]dx=
2[( _l)n
+2) 1] for ti = 2,3,4, ...
(
la
+ l)x]
7r1n
dx = O
al = ~ (" sin 2x dx = O
7r
Jo
bl = ~ (" sin 2 x
tt
Jo
f(x)
dx = 1
= sin z
2
L (1)n+1
2
1 n
00
f(x) =  + 7r 7r n=2
29. ao =
~(
7r
an = 2
7r
bn =
f (x)
("/2 X dx
Jo
(lo"'/2
o
+ i"
J"'/2
x cos nx dx
cos na:
(7r  x) dX)
'"
",/2
=~
2
(7r  x) cos nx dx
2 (n7r
= 22 cos 
n=l
n 7r
1) 1
+ (1 t+ 
ti
cos nx
4
n7r
L 2sin  sin nx
00
f (x) =
n=l
ti
2'"
=  2'"
30. ao= 1
tt
tt
(x7r)dx=
tt '"
an
tt r
bn = 1 211"( x 7r 11"
f (x)
7r +
=4
7r
)' sm
4 [ (1
2n.t:
t cos n7r]
2
n x dx = 2 ( 1)n+l   4. sm n7r
2
n
n27r
2
4 [
L
(1)
n=l n27r
00
508
tt
+ (1) n+l
1)
Exercises
n1f)
2 Slll
tt
2
+ 1,\dX
31. ao = (1 xdx
Jo
an =
. n X
sm
2
=~
2
~1f
1)
1,2.
. n1f
xsmxdx+
4.
n1f
1smxdx=2
bn=
ino
f (x)
3
00
4 (
n1f
)
nTi
=  + '" cos   1 cos  X
4
e:
n21f2
2
2
n=l
Slll
n21f2
32. ao = (11 dx
Jo
~ (4
()
f X = Z::
n=l
n tt
ino
. n1f
l'Slllxdx+
1,2
3
~
=4+~n21f2
n=l
fx()
f(x)
2 (
n+1
1)
n1f
1,\2  x) dx = ~2
Jo
n1f
2sm+2
2 n+1]' SlllXn1f
+ (1)
n1f
2
n1f
2
bn=
11.3
tt
n1f
2
(2x)sinxdx=+22sm2
n1f
(n1f
= 242 (COS ~
cos2+(1)
n+1)
+ (1 t+
4.
n1f
1f
1)
n1f
cos2x
00 ( 2
4
n1f)
n1f
= '"
sin sinx
e: n1f + n21f2
2
2
n=l
33. ao
=2
an =.2
bn = 2
+ x)
(\x2
Jo
(x2+x)
dx
=~
3
cos n1fX dx =
2(x2 + x)
11
2 'lo1
sin n1fX  (2x+ 1)sin n1fX dx
n1f
O
n1f O
4
+ _[(_l)n
= _(_l)n+1
n1f
f(x)
= 6+
f(x)
= '"
~
n=l
n31f3
00
cosn1fX
1]
L 22[3(
n=l n tt
00(4 _(_l)nr) .
n1f
2(x2+x)
n1f
lt
 1]cosnsrz
4
+ _[(_l)n
n31f3
1] sin nrz
509
= 222
2lo1
(2x + 1) costin dx
n1f o
+ 
1f
[3(l)n1]
Exercises 11.3
3"4
34. ao = Jo (2x  x ) dx =
2
an
bn
= Jor2 (2x
nt:
x2) cos xdx
2
la (2x O
2.
tu:
 x )sIn2xdx
lao
1 la
n7l"
bn
27r 2 .
=x SIn nx
7l" O
+ 2:
4
37l"2
00
n=l
r xdx
=~
4
x 2 cos nx dx = 2
n
7l"
1]cosx
= 1
f(x)
~7l"2
an
27r
16
1]
n37l"3[1 (1) ]
7l"
=.!.7l" Jor27rx2 dx =
7l"
n7l"
sinx
00
n=l
7l"
16
2: 33[1(_l)n]
f(x)
2 2[(lt+l
00
n=l
36. ao
= 3 + 2: 2'2[( lt+l
f(x)
35. ao
(4
dx = 47l"
n
47l")
 sinnx
n
2cosnx
n
= 7l"
7l" Jo
.:
an = 
7l"
bn
x cos 2nx dx = O
00
= 7l" + 2:
f(x)
37. ao
(1

=  1
sin 2nx
n=l
= 2la\X + 1) dx = 3
an=2
.
rl(x+
Jo
1)cos2n7l"xdx=
bn = 2
f(x)
38. ao
la (x + 1) sin 2n7l"xdx
O
3001
=2
= 
n7l'
2: 
n=l
sin 2n7l"x
n7l"
= 2la2 (2  x) dx = 2
510
Exercises
an
= 2 102 (2 
bn
= 2 10
f(x)
x) cos mrx dx
(2  x) sin mrx dx
= 1+
00
n=l
n7r
=O
=
n7r
L sinn7rx
39. We have
bn
= 2lo1r
10
5sin ntdt = [1(In
ntt
f(t) =
tt
so that
10[1  (l)n]
00
sin nt.
n7r
n=l
00
L En sin nt
n=l
and so
E  10[1  (_l)n
n  nst (O
1  n 2)'
Th
us
10
xp(t)
=
7r
1  (_l)n
00
L (
n=l n 10 
2) SlD nt.
40. We have
b
= 7r2 10O1(1 
t) sin n nt dt
2
= n7r
so that
f(t) =
00
n=l
n7r
L
sin nnt:
00
L En sin nnt
n=l
and so En
= n7r (O
1
2
2 2)'
 n t:
Thus
2001
xp(t) = 
L n (10 
7r n=l
511
2 2)
7r
sin ntit.
11.3
Exercises
11.3
41. We have
so that
21f2
f(t) =
(4)
00
3 + L ;z
cosnt.
n=l
:fA + f1
00
xp(t) =
Ancosnt
4' ;
f1 (1
+ 12xp = 6Ao .+
) cos nt = 3
21f2
(4)
+ J; ;z cos nt
00
An
00
+ 12
4'n2
1f2
and Ao
9'
42. We have
r1/2
ao = (1/2) Jo
r1/2
f(t)
= +
an = (1/2) Jo
= 2'
tdt
tcos2n1ftdt
= n21f2 [(_l)n
1]
so that
1
L
n=l
00
(1)n1
n
2 2
cos 2n1ft.
1f
1
and Ao = 24'
00
n=l
00
= +L
4
n=l
(1)n1
An = n21f2(12 _ n21f2) . Thus
1
xp(t) = 48
+ 1f2
f1
00
(l)n1
n2(12 _ n21f2) cos 2n1ft.
512
 1
(_l)n
2 2
1f
cos 2n1ft
Exercises
11.3
43. (a) The general solution is x(t) = el cos vT6t + e2sin vT6t + xp(t), where
1  (_l)n .
(
2) sin ni,
7["n=l n 10  n
10
00
xp(t) =  2::
The initial condition x(O) = O implies el + xp(O) = O. Since xp(O) = O, we have el = O and
x(t) = e2 sin vT6t + xp(t). Then x'(t) = e2vT6 cos vT6t + x~(t) and x'(O) = Oimplies
10 00 1 (_l)n
e2\l()"+  2::
2 cos 0= O.
7["n=l 10  n
Thus
vT6 00 1 (_l)n
e2 = ; 2:: 10 _ n2
n=l
and
10 00 l(l)n[l
2::
2
7["n=l 10  ti
1
]
 sin nt  117\ sin vT6t .
n
y 10
x(t) = 
(b) The graph is plotted using eight nonzero terms in the series expansion of x(t).
x
4
2
2
4
44. (a) The general solution is x(t) = el cos 4V3t + e2sin 4V3t + xp(t), where
7["2
xp(t)=+162::
18
00
2( 2
n=l n n 
48)cosnt.
E n2(n2 00
48) .
00
1)
x(t) = ( 1    162:: 2( 2
)
18
n=l n n  48
7["2
00
1
+ 16 2:: 2( 2
8) cosnt
18
n=l n ti  4
cos4..J3t + 
7["2 (
7["2)
00
1
=+
1cos4V3t+162::
2( 2 48)[cosntcos4..J3t1.
18
18
n=l n n 
513
Exercises
11.3
(b) The graph is plotted using five nonzero terms in the series expansion of x(t) ..
x
1.5
1
0.5
t
1
0.5
1
w(x)
mr
L 2wo (lt+ 1 sin L
x.
00
n=l
rttt
00
n7r
L En sin L x
then
n=l
= w(x)
gives
2wo( _l)n+l L 4
EIn57r5
Thus
46. We have
[n7r
2n7r]
2L/3
tvn
2wo
bn = wo sin  x dx = cos   cos L L/3
L
tit:
3
3
so that
w(x)
If we assume y(x)
00
2wo [n7r
L
cos n=l nt:
3
n7r
00
L En sin L x
then
n=l
514
2mr]
 cos 
tit:
sin 
x.
Exercses
Thus
11.3
= w(x) gives
4 00
n7r
2n7r
_ 2 WoL '" cos 3  COS3 . ti n
y (x )  5 L
5
sin L x.
El t: n=l
n
47. We note that w(x)is 2nperiodic and even. With p = tt we find the cosine expansion of
={
f(x)
\Ve have
ao = ~
r f(x)dx
n Jo
an
= 2
=~
r/
n Jo
wodx = Wo
= 2
7r
non
< x < n /2
n/2<x<n.
Wo
0,
f(x) cosnxdx
J7r /2
nn
= 2wo sin 2
.
Wocosnxdx
nn
Thus,
Wo
w(x) = 2
2wo
+
1 . nn
L
smcosnx.
n=l n
2
00
L~=l Ann4 cos nx and substituting into the differential equation, we obtain
Ely~4)(x)
+ kyp(x) =
kAo
2
+ L An(Eln4 + k) cosnx
00
n=l
Wo
2wo
=  +
1 . nn
L
sm n=l
2
00
cos nx.
ti
Thus
Ao
= Wo
and
_ 2wo sin(nn/2)
nn
n(Eln4 + k) ,
and
Wo 2wo 00 sin(nn /2)
yp(x) =  + (El 4 k) cosnx.
2k
n n=l n
n + .
then
= f(x)g(x)
= f(x)g(x)
= h(x)
and h is even.
(e) If f is even and 9 is odd and h(x)
h(x)
= f(x)g(x)
= f(x)g(x)
then
= f(x)[g(x)]
515
h(x)
Then y~4)(x) =
Exercises 11.3
and h is odd.
(d) Let h(x)
(f ) If j is even then
fa
La
j(x) dx =
=2
fa j(x)
Jo
dx.
_aaj(x)
dx = =
:a
O j(u)
loa j(u)
du + loa j(x) dx
du + loa j(x) dx = O.
jex) is even then j( x) = j(x). If j(x) is odd then j( x) =  j(x). Thus, if j(x) is both even
and odd, j(x) = j( x) =  j(x), and j(x) = O.
49. If
+ 00,
+ 00,
so we must have
C3
= a+ai,
mI
(C3
so we must have
cos ax
CI
= C4 = O. Thus, Yc
m2
= aai,
= C2 = O. We also expect
< 27f.
f (x)
2
1
2
1
2
3
516
y(x) to be
51. The graph is obtained by summing the series from n = 1 to 20. It appears that
X,o<x<7f
7r,
7r <
a+ai,
+ C4 sin oz ).
j(x) = {
m3 =
14
Exercises
= 1 to
11.4
O< x < 1
1 < x < 2.
1 x,
f(x) = { O,
f (x)
1.5
1.25
1
0.75
0.5
0.25
10
0.25
0.5
X,
f(x) =
x=7r
.,
7r
<
<
27r.
O,
x+l,
l<x<O
= { x + 1,
f(x)
O< x
<1
1 < x < 2.
O,
Exercises 11.4
1. For A
CI
cos ~
+ C2 sin ~
x.
Now
C2 =
O, so
+ y'(l)
CI(COS~
v'>..
+ C2~COS~X
~x
Xl,
= O or
~sin~)
cot ~
= V)...
517
= COS xnx
for n
1, 2, 3, ....
Using a CAS
Exercises
11.4
eigenfunctions cos 0.8603x, cos 3.4256x, cos 6.4373x, and cos 9.5293x.
2. For A
Now y' = el and the boundary conditions both imply el + e2 = O. Thus, A = O is an eigenvalue
with corresponding eigenfunction YO = x  1.
For A
> O we have
y = el cos ~ x + e2sin ~
and
y' = el~sin
~x
+ e2~cos~X.
+ e2~ = O
el cos ~
+ e2sin ~
= O
which gives
e2~cos~+e2sin~=0
or
tan~=~.
The eigenvalues are An = x;; where Xl, X2, x3, ... are the consecutive positive solutions of
tan /).. = /)...
 sin';>:';;' x (obtained by
positive eigenvalues are 20.1907, 59.6795, 118.9000, and 197.858 with corresponding eigenfunctions
4.4934 cos 4.4934xsin 4.4934x, 7.7253 cos 7.7253xsin 7.7253x, 10.9041 cos 10.9041xsin 10.9041x,
and 14.0662 cos 14.0662x  sin 14.0662x.
3. For A = O the solution of y" = O is y = elX + e2. The condition y'(O) = O implies el = O, so A = O
is an eigenvalue with corresponding eigenfunction 1.
For A
sin Nx
N x.
cosh Nx.
> O we have y = el cos ~ x + e2sin /).. x and y' = el/).. sin ~ x + e2/).. cos ~ x. The
condition y' (O) = O implies e2 = O and so y = el cos /).. x. Now the condition y' (L) = O implies
el/).. sin /).. L = O. For el =1= O this condition will hold when ~ L = mr or A = n27f2/ L2, where
For A
n = 1, 2, 3,
n
1, 2, 3,
4. For A
ci
cosh N x + e2sinh
y' = el N sinh
Nx+
518
Nx
e2N
cosh
N x.
Exercises
11.4
Using the fact that cosh x is an even function and sinh x is odd we have
y(L)
= clcosh(~L)
= Clcosh ~
+c2sinh(~L)
L  C2sinh ~
and
y'(L)
= cl~sinh(~L)
= Cl ~
+c2~cosh(~L)
L+
sinh ~
cosh ~
C2~
L.
L  C2sinh ~
L = Clcosh
or
v=>: L = O
2C2 sinh
and
or
= o.
2Cl~sinh~L
Since
y
R L =J o, Cl = C2= O and
= O.
For A = O we have
y
= ClX
y'
From y(L)
= Cl
= y(L) we obtain
=ci l.
Then Cl = O and y
For A
+ C2
+ C2= clL + C2
> O we have
y = ci
y' = 
COS
ci
C2,j)._
cos ,j)._ x.
or
2C2 sin
,j)._ L = O.
519
= O.
Exercises 11.4
Thus, if
el
= O and
e2
i= O,
V).,L=mr
n; x, for n
condition implies
2q V)., sin V)., L
If
e2 = O and el
= O.
i= O,
n21f2
A = V' n
= 1,2,3, ... ,
n1f
where cot
A = Fn.
Thus
=~
=~
[1 + 2~
=~
[1 + ~
sin
=~
[1 + ~
(sinFn)
(2 sin
r:
11 = ~
cos
(1 +
~ sin 2>:;')
2y An
>:;.) ]
dx
Fn (sin Fn)]
=~
= ~2
(x
2y
=~
[1
2~
=~
[1 ~
=~
[1 ~ ( Fn cos
o= ~
2
~sin2Fnx)
x,
11
(2 sin .:
tan
= An. Thus
(1 + sin Fn) .
cos
(1
dx
~sin2Fn)
2y An
>:;.) ]
520
= ~(
1 + cos Fn) .
2
Exercises
7. (a)
11.4
If,\::; O the initial conditions imply y = O. Por .\ > O the general solution of the CauchyEuler
differential equation is y = el cos(.J:\ lnx) + e2sin(.J:\ lnx). The condition y(l)
= O implies
n1T,
2, 3, ....
n = 1,
d [xy '] + y
dx
x
= O.
JI
i( 1
VI  4,\). When ,\
= O the
and el + e2e2 = O. Since the determinant of the coefficients is not O, the only solution of this
homogeneous system is el = e2 = O, in which case y = O. Similarly, if O
solution is
y' = ele~(1+v'14A)x + e2e~(Iv'14A)x.
< .\ <
!'the general
In this case the initial conditions again imply el = e2 = O, and so y = O. Now, for A
> ~, the
= c2e1
sin 2)4'\  1 = O
= n1T for n
= 1, 2, 3, .. , . Thus,
eigenfunctions
Cxj2sin~x.
2
(b) The selfadjoint form is