100 Math 1

100
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
. .
. .
. .
. .
. .
(C)
. .
. . . . .
. .
. .
A . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
991
. 991
. 9911
. 9916
. 9923
. 9931
. 9942
. 9949
. 9959
. 9965
. 9970
. 9977
. 9983
. 9990
. 99100
. 99108
100
991
1. 100
1. (1) Solve x3 y + xy y = 0

(2) Solve (y x x2 + y 2) dx + (x y x2 + y 2 )dy = 0. (15%)
100
(1) y = xm ODE
m(m 1)(m 2) + m 1 = 0 m = 1 , 1 , 1
y(x) = c1 x + c2 x ln x + c3 x(ln x)2

(2) ODE
3
1
d{xy (x2 + y 2 ) 2 } = 0
3
ODE
3
1
xy (x2 + y 2 ) 2 = c
3
1
and y(0) = 1, y (0) = 0; use Laplace
2
s +1
100
transform to solve ty + y + ty = 0.
2. (15%) Knowing L {J0 (t)} =
ODE Laplace
L {ty(t) + y (t) + ty(t)} = 0
d 2
d
{s Y (s) sy(0) y (0)} + sY (s) y(0) Y (s) = 0
ds
ds
L {y(t)} = Y (s) ,
2sY (s) s2
dY (s)
d
+ y(0) + sY (s) y(0) Y (s) = 0
ds
ds
992
http://www.superyu.idv.tw
(s2 + 1)
dY (s)
+ sY (s) = 0
ds
dY (s)
s
+ 2
Y (s) = 0
ds
s +1
dY
s
= 2
ds
Y
s +1
1
ln |Y (s)| = ln |s2 + 1| + c
2
Y (s) =
c
s2 + 1
lim sY (s) = lim
c = 1 , Y (s) =
cs
= c = y(0) = 1
s2 + 1
1
, ODE
s2 + 1
y(t) = L 1 {Y (s)} = L 1 {
1
} = J0 (t)
+1
s2
3. (10%)
100
(1) Write the 3 3 matrix of the geometric transformation representing the z-axis
counterclockwise rotation (i.e., the axis of rotation perpendicular to the x-y
plane).
(2) Find the eigenvalues and eigenvectors of this 3 3 transformation matrix.
(1)

x
cos

y = sin
z
0
sin
cos
0

0
x

0y
1
z
1.
993
cos
A = sin
0
sin
cos
0
0
1
(2)
det(A I) = (1 )(2 2 cos + 1) = 0
A = 1 cos i sin = 1 (A I)X = 0

cos 1
sin
0
0
x1

sin cos 1 0 x2 = 0
x3
0
0
0
0

0

X = c1 0
(c1 = 0)
= cos + i sin (A I)X = 0
i sin
sin
0
sin
i sin
0

1

X = c2 i
0

0

0 x2 = 0
x3
0
0
0
x1
(c2 = 0)
= cos i sin
1

X = c3 i
0
(c3 = 0)
994
4. Consider an elastic string of length L, xed at its ends on the x axis at x = 0 and
x = L. Its displacement function satises :
2
2y
2 y
=c
for 0 < x < L , t > 0
t2
x2
y(x , 0) = f (x) for 0 x L,
y
(x , 0) = g(x) for 0 x L.
t
(1) (8%) For zero initial velocity, which of the functions listed below gives the
correct displacement (Justication of your answer is required to ge credit.)
nx
nct
n
2
d } sin
cos
{
f () cos
(a) y(x , t) =
L n=1 0
L
L
L
nx
nct
2
n
d } sin
cos
(b) y(x , t) =
{
f () sin
L n=1 0
L
L
L
nx
nct
2
n
d } sin
sin
(c) y(x , t) =
{
f () cos
L n=1 0
L
L
L
nx
nct
2
n
d } sin
sin
(d) y(x , t) =
{
f () sin
L n=1 0
L
L
L
6x
, determine the displace(2) (6%) For zero initial velocity and f (x) = 95 sin
L
ment function.
6x
, the displacement is
(3) (6%) For zero initial displacement and g(x) = 95 sin
L
3x
3ct
95
sin
sin
. What is the displacement if
y(x , t) =
3c
L
L
6x
?
f (x) = g(x) = 95 sin
L
1
2
(4) (5%) For zero initial velocity and f (x) = H(x L) H(x L), sketch
3
3
L
y(x , ).
6c
L
100
(5) (5%) Following (4), sketch y(x , ).
2c
1.
995
(1) y(0 , t) = y(L , t) = 0,

sin
y(x , t) =
n=1
nx
,
L n=1
an (t) sin
nx
L
PDE

n=1

nx
n2 2
nx
= c2
an (t)( 2 ) sin
L
L
L
n=1
an (t) sin
{an (t) + (
n=1
nc 2
nx
) an (t)} sin
=0
L
L
an (t) + (
nc 2
) an (t) = 0
L
an (t) = An cos
nct
nct
+ Bn sin
L
L
PDE
y(x , t) =
{An cos
n=1
nct
nct
nx
+ Bn sin
} sin
L
L
L
I.C.
y(x , 0) = f (x) =

n=1
2
An =
L
An sin
f () sin
0
nx
L
n
d
L

y
nc
nx
(x , 0) = 0 =
sin
Bn
t
L
L
n=1
Bn = 0 ,
2
y(x , t) =
{
L n=1
f () sin
0
nct
nx
n
d} cos
sin
L
L
L
(b)
(2) g(x) = 0, Bn = 0,
y(x , 0) = 95 sin

6x
nx
An sin
=
L
L
n=1
A6 = 95 , An = 0,
y(x , t) = 95 cos
6x
6ct
sin
L
L
(1)
996
(3) , ,
y(x , 0) = f (x) = 95 sin

6x
nx
=
An sin
L
L
n=1
A6 = 95 , An = 0,

y
nc
6x
nx
Bn
(x , 0) = 95 sin
=
sin
t
L
L
L
n=1
B6
6c
95L
= 95, B6 =
, Bn = 0
L
6c
y(x , t) = {95 cos
6ct
6x
6ct 95L
+
sin
} sin
L
6c
L
L
(4) y(x, , t) = (x + mt) PDE ,

c2 m2 = 0 m = c
PDE
y(x , t) = (x + ct) + (x ct)
(1)
y(x , 0) = f (x) f (x) = (x) + (x)
(2)
y
(x , 0) = g(x) g(x) = c (x) c (x)
t
(3)
{(3) } d
1
c
g() d = (x) (a) (x) + (a)
(4)
(2) + (4)
1
f (x) +
c
g() d = 2(x) (a) + (a)
1
1
(x) = {f (x) +
2
c
g()d + (a) (a)}
(2) (4)
1
f (x)
c
x
a
g() d = 2(x) + (a) (a)
(5)
1.
997
1
1
(x) = {f (x)
2
c
g()d (a) + (a)}
(6)
(5) (6) (1) , DAlembert

y(x , t) = (x + ct) + (x ct)
1 x+ct
1
{f (x + ct) +
g()d + (a) (a)}
=
2
c a
1 xct
1
+ {f (x ct)
g()d (a) + (a)}
2
c a
1 x+ct
1
1 xct
{f (x + ct) + f (x ct)} +
=
g()d
g()d
2
2c a
2c a
1
1 x+ct
=
g()d
{f (x + ct) + f (x ct)} +
2
2c xct
1
3
2
3
f (x) = H(x L) H(x L) g(x) = 0,

L
2L
L
2L
1
) + H(x ct ) H(x ct
)}
y(x , t) = {H(x + ct ) H(x + ct
2
3
3
3
3
y(x ,
1
L L
L 2L
L L
L
) =
{H(x + ) H(x +
) + H(x )
6c
2
6
3
6
3
6
3
L 2L
)}
H(x
6
3
L
L
L
5L
1
{H(x ) H(x ) + H(x ) H(x
)}
=
2
6
2
2
6
0 ; 0 x < L/6
1
; L/6 x < 5L/6
=
0 ; 5L/6 x L
(5)
y(x ,
1
L L
L 2L
L L
L
) =
{H(x + ) H(x +
) + H(x )
2c
2
2
3
2
3
2
3
L 2L
)}
H(x
2
3
L
L
5L
7L
1
{H(x + ) H(x ) + H(x
) H(x
)}
=
2
6
6
6
6
; 0 < x < L/6
=
0 ; L/6 x < 5L/6
; 5L/6 x < L
2
998
5. (15%) Write down the answers to the following questions : ( Derivations are nor
required. )

cos
n d over a circle C of
(1) Let (r , ) = 2 . Evaluate the line integral
r
C
radius 2 centered at the origin. (

n denotes unit vector normal to the contour
of the circle C. )
(2) Let (r , ) satisfy the 2 - D Laplace equation (i.e., 2 = 0 ); and
= cos2 sin2 ,
= 2 cos 2 along the contour of a unit circular disk S.
r
dA over the circular disk S.

Evaluate the surface integral
S

y
x

F d
r
i + 2
j be a 2 - D vector eld. Evaluate
(3) Let F = 2
2
2
x +y
x +y
C
over a closed contour C given by :
100
n =
e r = cos i + sin j , d = 2 d,
(1) C : x2 + y 2 = 4,

cos
n d =
( 2 )(cos i + sin j )2 d
r
C
C
2
1
=
(cos2 i + cos sin j )2 d
2
2
0
=
2
(2)
() dA =
( + 2 ) dA ( S 2 = 0 )
dA
=
S
1.
999

=

n ds
er+
e )
e r d
r
r
n =
e r , ds = d)
( S C, C
d
=
r
C
2
=
(cos2 sin2 )(2 cos 2) d
0
2
= 2
cos 2 cos 2 d = 2
(3) (x0 , y0 ) C , C , (x0 , y0 ) = (0 , 0),
x dx
y dy

F dr =
( 2
+ 2
)
2
x +y
x + y2
C
C
1
1
d(x2 + y 2 )
=
2
2
x
+
y
2
C
1
d(ln |x2 + y 2|)
=
2 C
(x0 , y0 )
1
2
2
ln |x + y |
=
(x0 , y0 )
2
= 0
9910
6. (15%) Let z = x+iy denote the complex variable, z = xiy the complex conjugate
of z, and f (z) is complex function. Answer the following questions. ( Derivations
are not required
)
z
d z over C : |z| = 1.
(1) Evaluate
C z (i/2)
(2) Let f (z) be analytic on the upper-half of z-plane and |f (z)| 0 as |z| .
xi
, nd the function f (z).
If on the real axis, f (z) takes the form 2
x +1
hint : Apply Cauchy integral formula.
z
about z = 1 is
(3) If the Laurent series expansion of f (z) =
(z i)(z + 1)2

an (z + 1)n , nd
an =?
100
denoted by
n=
n=
1
1
1
(1) C : |z| = 1, z = ei , = 0 2 , z = ei = , d z = d( ) = 2 dz ,
z
z
z
1

z
2
1
z
dz =
dz
( 2 )dz =
2
1
i
z
z (iz 2)
C z (i/2)
C
C
z 2
2
, C z = 0 2 pole,
f (z) = 2
z (iz 2)
d
2
i
(
)=
z0 dz iz 2
2
Resf (0) = lim

C
z
dz =
z (i/2)
(2)
2
i
dz = 2 iResf (0) = 2 i( ) =
2)
2
z 2 (iz
z i ; z = i
f (z) = z 2 + 1
;z=i
f (z) , C : |z i| = ( 0+) ,

1
1
1
f (z)
1
dz =
dz = Resf (i) =
= f (i) =
2
2i C z i
2i C z + 1
2i

an (z + 1)n
=
an = f (0) = 0
(3)
n=
z=0
n=
2.
9911
2. 100
1. (20%) Given the Bessels equation x2 y + xy + (x2 2 )y = 0 and the solution
can be expressed as y(x) = AJ (x) + BY (x). Find the general solution ( in
terms of the Bessel
function) of the equation my + ket y = 0 with the indicated
2 k t/2
e
.
100
substitution x =
m

dy
k t/2
dy dx
dy
)
=
=
(
e
dt
dx dt
dx
m

d2 y
k t/2
d dy
{ (
e
=
)}
2
dt
dt dx
m

k t/2
k t/2
d2 y dx
dy d
(
e
{
e
=
)+
}
2
dx dt
m
dx
dt
m

d2 y
k t/2 2 dy
k t/2
e
e
=
(
) + {
}
2
dx
m
dx 2 m
2 2 d2 y 2 dy
x
x
+
=
4
dx2
4 dx
ODE
m{
2 2 d2 y 2 dy
x
x } + ket y = 0
+
2
4
dx
4 dx
k
2 2 d2 y 2 dy
x
x } + et y = 0
+
2
4
dx
4 dx
m
2 2 d2 y 2 dy
2 2
x
x
}
+
x y=0
+
4
dx2
4 dx
4
x2
d2 y
dy
+ x + x2 y = 0
2
dx
dx
2
y(x) = c1 J0 (x) + c2 Y0 (x) = c1 J0 (
k t/2
2
e
) + c2 Y 0 (
m
k t/2
e
)
m
9912
2. (20%) If f (x) is a periodic function with period of T , prove that

1
T
2
1
, a0 =
where w0 =
T
T
2
an =
T
T /2
T /2
f 2 (x) dx a20 +
k
1
(a2 + b2n )
2 n=1 n
T /2
f (x) dx ,
T /2
T /2
2
f (x) cos nw0 x dx ; bn =
T
T /2
T /2
T /2
f (x) sin nw0 x dx

100
f (x) T ,
f (x) = a0 +
(an cos
n=1
2nx
2nx
+ bn sin
)
T
T
T /2
< f (x) , f (x) > =

T /2
f 2 (x) dx
2nx
2nx
+ bn sin
),
T
T
n=1

2mx
2mx
+ bm sin
)
a0 +
(am cos
T
T
m=1

2nx
2nx
2
, cos
>
= a0 < 1 , 1 > +
a2n < cos
T
T
n=1
2nx
2nx
, sin
>
+b2n < sin
T
T

T
T
= a20 T +
(a2n + b2n )
2
2
n=1
1
T
1
T
a0 +
T /2
T /2
T /2
T /2
(an cos
f 2 (x) dx = a20 +
1
(a2n + b2n )
2 n=1
f 2 (x) dx a20 +
k
1
(a2 + b2n )
2 n=1 n
2.
9913
2u
2u
3. (20%) Consider the 1D wave equation
=
and the boundary conditions
t2
x2

u
u
are specied as
= 0 and
= 0. If the initial conditions are given by

x x=0
x x=L
7x
u
3x
+ 5 cos
and
u(x , 0) = 1 + 2 cos
= 0. Find the solution of u(x , t).
L
L
t t=0
100

u
u
nx
,
(0 , t) = 0
(L , t) = 0 , 1 , cos
x
x
L n=1
u(x , t) = a0 (t) +

n=1
an (t) cos
nx
L
PDE
a0 (t) +

n=1
an ](t) cos
a0 (t) +

nx
nx
n2 2
=
) cos
an (t)(
L
L
L
n=1
{an (t) + (
n=1

a0 (t) = 0
a (t) + ( n )2 an (t) = 0
n
L
u(x , t) = A0 + B0 t +
n 2
nx
) an (t)} cos
=0
L
L
a0 (t) = A0 + B0 t
an (t) = An cos nt + Bn sin nt
L
L
{An cos
n=1
u(x , 0) = 1 + 2 cos
nx
nt
nt
+ Bn sin
} cos
L
L
L

7x
3x
nx
+ 5 cos
= A0 +
An cos
L
L
L
n=1
A0 = 1 A3 = 2 A7 = 5, An = 0,

nx
u
n
(x , 0) = 0 = B0 +
cos
Bn
t
L
L
n=1
B0 = 0 Bn = 0,
u(x , t) = 1 + 2 cos
3x
7t
7x
3t
cos
+ 5 cos
cos
L
L
L
L
9914
4. (20%) For an Eigenvalues problem Ax = x. prove that the eigenvalues are real
100
if A is a Hermitian matrix.
Hermitian A x ,
Ax = x
(1)
(1) ,
(Ax) = (x)
x A = x
(2)
A Hermitian , A = A , (2) ,
x A = x
x
x Ax = x x
x x = x x
( )x x = 0
x x = 0 , =
5. (20%) Consider the following two initial values problems :

(1) m
y + cy + ky = (t) I.C. y(0) = y(0)
= 0, (t) : Dirac Delta function.

(2) m
y + cy + ky = r(t) I.C. y(0) = y(0)
=0.
IF h(t) is the solution of problem (1), show that the solution of problem (2)
can be expressed by :
y(t) =
h(t )r( ) d
100
2.
9915
(1)
m
y + cy + ky = (t)
LT
m{s2 Y (s) sy(0) y(0)}
+ c{sY (s) y(0)} + kY (s) = 1
L {y(t)} = Y (s),
Y (s) =
1
ms2 + cs + k
h(t) = L 1 {Y (s)} = L 1 {
ms2
1
}
+ cs + k
(2)
m
y + cy + ky = r(t)
LT
m{s2 Y (s) sy(0) y(0)}
+ c{sY (s) y(0)} + kY (s) = R(s)
L {r(t)} = R(s),
Y (s) =
ms2
1
R(s)
+ cs + k
1
R(s)}
y(t) = L 1 {Y (s)} = L 1 { 2
ms + cs + k
t
= h(t) r(t) =
h(t )r( )d
0
9916
3. 100
1. (30%) Suppose A is a real symmetric square matrix such that A2 = A.
(a) (5%) Find the determinant of A (i.e.; det A ).
(b) (5%) Find the eigenvalues of A.
(c) (5%) Let x be a column vector and the symbol |x| denote it vector length.
Find (Ax)T (x Ax) (i.e.; the inner product of two column vectors Ax and
(x Ax).
(d) (5%) Show that |Ax| 1 for any column vector x with |x| = 1.
(e) (10%) Suppose A here denote a real symmetric 3 3 matrix. Let t, m, and n
be its orthonormal eigenvectors such that
Am = m , An = n , At = 0
|m| = 1 , |n| = 1 , |t| = 1 , tT m = tT n = mT n = 0
1 0 0
Let P = [ m | n | t ] and = 0 1 0 be the 3 3 matrices containing the

0 0 0
eigenvectors and eigenvalues of A. Suppose t = (t1 , t2 , t3 )T . Find P P 1 in
100
terms of t1 , t2 , t3 .
(a) A2 = A , det(A2 ) = {det(A)}2 = det(A), det(A) = 0 det(A) = 1

(b) A , v,
Av = v
A
A2 v = Av Av = Av ( 1)Av = 0
Av = 0 1 = 0, A = 0 = 1
(c) A , AT = A, A2 = A, AT A = A2 = A,
(Ax)T (x Ax) = xT AT (x Ax) = xT AT x xT AT Ax = xT Ax xT Ax = 0
3.
9917
(d) |x| = 1, xT x = 1,
|Ax|2 =< Ax , Ax >= (Ax)T (Ax) = xT AT Ax = xT A2 x = xT Ax
(1)
Rayleighs
min
xT Ax
= xT Ax < max
xT x
(2)
A = 1, (1) (2)
|Ax|2 = xT Ax 1 |Ax| 1
(e) P = [ m | n | t ],
1 0 0
mT
mT
P P 1 = [ m | n | t ] 0 1 0 nT = [ m | n | 0 ] nT = m mT + n nT
tT
tT
0 0 0
2. (a) (8%) Given a 1st order system ODE :

y (x) = Ay(x) , x > 0

1
2 1
y(0) =
, where A =
. Solve y(x).
1
0 2
(b) (6%) Given a 1st order ODE :
y (x) =
y ln y
, y(0) = 1
x
Solve y(x).
(c) (8%) Find the general solution of the following 2nd order ODE :
x2 y xy + y = 0
(d) (10%) Given a 2nd order ODE :
y (x) + k 2 y(x) = f (x) , x > 0
where k is a real number, and f (x) is an arbitrary piecewisely continuous real
function. Solve y(x).
100
9918
(a) det(A I) = 0, = 2 2 , = 2 (A I)V = 0

0 1
v1
0
=
0 0
0
v2

1
V = c1
0
(A I)V2 = V2 ,
0 1
0 0
v1 = 0 v2 = 1,

1
(c1 =
0) V1 =
0

v1
v2
0
V2 =
1

1
=
0

1 2x
0
1
e + c2 (
y(x) = c1
+x
) e2x
0
1
0

1
0
1
+ c2
y(0) =
= c1
0
1
1

c1 = 1 c2 = 1,

2x
1 2x
0
1
e xe2x
2x
y(x) =
e (
+x
)e =
e2x
0
1
0
(b)
dy
1
= dx
y ln y
x
ln | ln y| = ln |x| + c1
ln y = c1 x , y(0) = 1 ,
(c) y = xm ODE ,
m(m 1) m + 1 = 0 m2 2m + 1 = 0
m=11,
y(x) = c1 x + c2 x ln x
(d) ODE LT ,
s2 Y (s) sy(0) y (0) + k 2 Y (s) = F (s)
3.
9919
L {y(x)} = Y (s) L {f (x)} = F (s),

Y (s) =
s y(0)
y (0)
F (s)
+
+ 2
2
2
2
2
s +k
s +k
s + k2
y(x) = L 1 {Y (s)}
y (0)
= y(0) cos kx +
sin kx +
k
y (0)
sin kx +
= y(0) cos kx +
k
1
f (x) sin kx
k
1 x
f ( ) sin k(x ) d
k 0
3. The height of a hill(in meter) is given by

h(x , y) = 10(2xy 3x2 4y 2 18x + 28y + 12)
where y is the distance(in km) north, x the distance east of Heaven city.
(a) (3%) Where is the top of the hill located ?
(b) (2%) How high is the hill ?
(c) (3%) How steep is the slope at a point 1 km north and 1 km east of Heaven
city, and in what direction ?
(d) (2%) Calculate h and h.
100
(a)
= 10(18 6x + 2y) = 0
x
h
= 10(28 + 2x 8y) = 0
y
x = 2 y = 3 , hill x = 2 km y = 3 km
(b) hill h(2 , 3) = 720 m
(c)
h =
h
h
i +
j = {10(18 6x + 2y)} i + {10(28 + 2x 8y)} j
x
y
9920
h(1 , 1) = 220 i + 220 j , (1 , 1) slope
|h(1 , 1)| = 220 2

i + j
2
(d) h =
2h 2h
+
= 60 80 = 140
x2 y 2
h = 0

{2xy dx + (ex + x2 )dy} along the curve C, where
4. (10%) Evaluate the integral

C
C is the boundary of the triangle with vertices (0 , 0), (1 , 0) and (1 , 1), along
clockwise direction.
100
C D , Greens

(e + x2 )
(2xy)} dxdy
y
D x
x=1
y=x
=
ex dy dx
{2xy dx + (e + x )dy} =
x
x=0
x=1
x=0
y=0
x=1

xex dx = (xex ex )
=1
x=0
3.
9921
5. The one dimensional wave equation

2
2u
2 u
=
c
t2
x2
describes the transverse displacement, u(= u(x , t)), of an elastic stretched string,

where c = T / is the wave speed, with T the tension in the string, and the
density of the string.
(a) (15%) Solve the equation subject to boundary conditions u(0 , t) = u( , t) = 0
u
(x , 0) = g(x). Here is the length
and initial conditions u(x , 0) = f (x),
t
of the string.
(b) (5%) Describe the physics of the problem, including the equation, the boundary conditions, the initial conditions and the solution.
State also the
assumptions for deriving the wave equation.
100
(a) u(0 , t) = u( , t) = 0,
u(x , t) =
n=1
an (t) sin
nx

PDE

n=1
an (t) sin

nx
n2 2
nx
= c2
an (t)( 2 ) sin

n=1
{an (t) + (
n=1
nc 2
nx
) an (t)} sin
=0

an (t) + (
an (t) = An cos
u(x , t) =
{An cos
n=1
nc 2
) an (t) = 0

nc t
nc t
+ Bn sin

nc t
nc t
nx
+ Bn sin
} sin

u(x , 0) = f (x) =

n=1
An sin
nx

(1)
9922
2
An =

f (x) sin
0
ut (x , 0) = g(x) =
2
nc
)=
Bn (

n=1
Bn (
nx
dx

nx
nc
) sin

g(x) sin
0
nx
dx

An Bn (1)
(b) PDE PDE, , f (x),
g(x), :
(1) , u(x , t)
(2) , bending
(3)
(4) (u )
4.
9923
4. 100
1. 2(3x + 1)2 y + 21(3x + 1)y + 18y = 0. (10%)
100
y = (3x + 1)m , ODE

2 32 m(m 1) + 21 3m + 18 = 0
6m2 + 15m + 6 = 0 m = 2 ,
1
2
y(x) = c1 (3x + 1)2 + c2 (3x + 1) 2
2. Laplace Transformation (15%)

y + 6y + 8y = e3t e5t , y(0) = 0 , y (0) = 0
100
ODE LT
s2 Y (s) sy(0) y (0) + 6{sY (s) y(0)} + 8Y (s) =
1
1
s+3 s+5
L {y(t)} = Y (s) ,
(s2 + 6s + 8)Y (s) =
1
2
1
=
s+3 s+5
(s + 3)(s + 5)
2
(s + 2)(s + 4)(s + 3)(s + 5)
1
1
1
1
=
3(s + 2) s + 3 s + 4 3(s + 5)
Y (s) =
9924
1
1
y(t) = L 1 {Y (s)} = e2t e3t + e4t e5t
3
3
3. yex dx + (2y + ex )dy = 0. (10%)
100
d(yex + y 2 ) = 0 , ODE
yex + y 2 = c
4. y 6y + 12y 8y =
2x e2x . (15%)
100
(1) : y = emx ODE

m3 6m2 + 12m 8 = 0 m = 2 , 2 , 2
yh (x) = c1 e2x + c2 xe2x + c3 x2 e2x

(2) :
1
1
2x e2x = e2x 3 2x
3
(D 2)
D

2 2x 7/2
8
e x
2x dxdxdx =
= e2x
105
yp (x) =
(3) : y(x) = yh (x) + yp (x)
4.
9925
5. , (x , y) T (x , y) = 5 + 2x2 + y 2
(1) ? (1%)
(2) (4 , 2) ,
? (3%)
(3) (4 , 2)
r (t) = x(t) i + y(t) j (6%)

100
(1) (0 , 0)
(2)
T (x , y) = 4x i + 2y j
(3)
16 i + 4 j
1
T (4 , 2)

=
= (4 i + j )
|T (4 , 2)|
4 17
17
d
r
dx
dy
=
i +
j 4x i + 2y j
dt
dt
dt
dx
dy
= 4mx ,
= 2my
dt
dt
x(t) = c1 e4mt , y(t) = c2 e2mt
9926
t = 0 x(0) = 4 = c1 y(0) = 2 = c2 ,
r (t) = x(t) i + y(t) j = 4e2t i + 2et j
6.
(1) u(r , )
(3%)
(2) u(r , ) (12%)
100
(1)
2 u(r , ) =
1 2u
2 u 1 u
+
+
=0
r 2
r r r 2 2
u( , ) = u0 , ur (0 , ) = 0 , u(r , 0) = 0 , u(r , ) = 0
2
4
(2) u(r , 0) = 0 , u(r , ) = 0 ,

4
u(r , ) =

n=1
an (r) sin(4n)
(1)

n=1
an (r) sin(4n) +
1
1
an (r) sin(4n) + 2
an (r)(16n2 ) sin(4n) = 0
r n=1
r n=1
{r 2 an (r) + r an (r) 16n2 an (r)} sin(4n) = 0
n=1
(1)
4.
9927
r 2 an (r) + r an (r) 16n2 an (r) = 0
Euler-Cauchy ,
an (r) = An r 4n + Bn r 4n
u(r , ) =
{An r 4n + Bn r 4n } sin(4n)
(2)
n=1

1
1
1
u( , ) = u0 =
{An ( )4n + Bn ( )4n } sin(4n)
2
2
2
n=1
1
1
8
An ( )4n + Bn ( )4n =
2
2
ur (1 , ) = 0 =
u0 sin(4n) d =
2 u0 (1 + cos n)
n
(3)
{4n An 4n Bn } sin(4n)
n=1
4n An 4n Bn = 0
(3) (4)
An = Bn =
21+4n u0 (1 + cos n)
(1 + 28n )n
An Bn (2)
7.
(1) I = [i1 (t) i2 (t)]T i1 (t) i2 (t)
dI
= AI + BE (3%)
,
dt
(2) R1 = 8 , R2 = 3 , L1 = 1H , L2 = 1H , E(t) = 100 sin t V, i1 (0) = 0
i2 (0) = 0 (variation of parameters)

100
(12%)
(4)
9928
(1)
i1 = i2 + i3
di2
di1
+ L2
=E
i1 R1 + L1
dt
dt
i3 R2 L1 di2 = 0
dt
di1
= E i1 R1 i3 R2 = E i1 (R1 + R2 ) + i2 R2
L2
dt
di
L1 2 = i3 R2 = (i1 i2 )R2
dt
dI
= AI + BE
dt

A=
(R1 + R2 )/L2
R2 /L2
R2 /L1
R2 /L1
(1)

, B=
1/L
(2) L1 = L2 = 1, R1 = 8 R2 = 3 E = 100 sin t (1)

11 3
1
A=
, B=
3
3
0
det(A I) = 0, = 2 12, = 2 (A I)X = 0

9 3
3 1

X1 =
x1
x2

0
=
0

1
3
= 12 (A I)X = 0

1 3
3 9

x1
x2

0
=
0
4.
9929
3
X2 =
1
ODE ODE (fundamental matrix)

2t
Q = [X1 e
12t
X2 e
]=
e2t
3e12t
3e2t
e12t
Ih = QC, C 2 1 , Q = AQ
(b) Ip : Ip = QY ,
y1 (t)
Y =
y2 (t)
Ip = Q Y + QY , (1) ( ODE)

Q Y + QY = AQY + BE QY = (Q AQ)Y + BE QY = BE
2t

e
100 sin t
3e2t
1
dt
=
Q BE dt =
10 3e12t e12t
0
2t
2t

4e
sin
t
2e
cos
t
2t
10e sin t
6
dt = 72 12t
=
e sin t e12t cos t
30e12t sin t
29
29

Ip = QY =
2t
e
3e2t
12t
3e
e12t
4e2t sin t 2e2t cos t
72 12t
6
e sin t e12t cos t
29
29
332
76
sin t
cos t
29
29
=
276
168
sin t
cos t
29
29
332
76

12t
sin t
cos t
e2t
3e
29
29
I = Ih + Ip = c1
2t + c2
12t +
276
168
3e
e
sin t
cos t
29
29
76

1
3
i1 (0)
0
29
=
+ c2
+
= c1
I(0) =
168
3
1
i2 (0)
0
29

9930
c1 = 2 c2 =
6
,
29

I=2
2t
2t
3e

+
12t
6 3e
29 e12t
332
76
sin t
cos t
29
29
+
276
168
sin t
cos t
29
29
8. F = (16y + sin x2 ) i + (4ey + 3x2 ) j C

, C C1 C2 C3 (10%)
100

C
F d
r =
(16y + sin x2 )dx + (4ey + 3x2 )dy
(16y + sin x2 )}dxdy

{ (4ey + 3x2 )
=
x
y

R
(6x + 16)dxdy = (6x + 16)A ( A R )
=
R
= 16
= 4
4
5.
9931
5. 100
1. Please nd the general solution of the following ordinary dierential equation
(O.D.E.s):
(a) (6%) (x sin )d + (x3 2x2 cos + cos )dx = 0
100
(b) (6%) (xD 4 + D3 )y = 150x4 .
(a)
M = x3 2x2 cos + cos

N = x sin
= 2x2 sin sin
= sin
x
M
N
x = 2x sin sin sin = 2x 2

N
x sin
x
2
ex
2
I = exp{ (2x ) dx} = ex 2 ln |x| = 2
x
x
ODE
2
ex
ex
2
2
sin d + xex 2ex cos + 2 cos theta)dx = 0
x
x
1 2
ex
cos + ex ) = 0
d(
x
2
ODE
1 2
ex
cos + ex = c
x
2
(b)
x4 y (4) + x3 y = 150x7
x = et , t = ln x,
x3 y = Dt (Dt 1)(Dt 2)y , x3 y (4) = Dt (Dt 1)(Dt 2)(Dt 3)y
9932
Dt =
d
, ODE
dt
Dt (Dt 1)(Dt 2)(Dt 3)y + Dt (Dt 1)(Dt 2)y = 150e7t
Dt (Dt 1)(Dt 2)2 y = 150e7t

(1) : y = emt (1)
m(m 1)(m 2)2 = 0 m = 0 , 1 , 2 , 2
yh = c1 + c2 et + c3 e2t + c4 t e2t = c1 + c2 x + c3 x2 + c4 x2 ln x
(2) :
1
150
1 7
7t
7t
x
150e
=
e
=
Dt (Dt 1)(Dt 2)2
7 6 52
7
1
(3) : y(x) = yh + yp = c1 + c2 x + c3 x2 + c4 x2 ln x + x7
7
yp =
2. (10%) Solving the following ODE by Laplace transform.

1 if 1 < t < 2

y + 3y + 2y = r(t) , r(t) =
0 otherwise
100
y(0) = y (0) = 0.

r(t) =
1 if 1 < t < 2
= H(t 1) H(t 2)
0 otherwise
1 s 1 2s
L {r(t)} = e e
s
s
ODE LT

L {y + 3y + 2y} = L {r(t)}
1
1
s2 Y (s) sy(0) y (0) + 3{sY (s) y(0)} + 2Y (s) = es e2s
s
s
L {y(t)} = Y (s),
1
1
(s2 + 3s + 2)Y (s) = es e2s
s
s
(1)
5.
9933
1
1
es
e2s
s(s + 1)(s + 2)
s(s + 1)(s + 2)
1
1
1
1
1
1
= {
+
}{
+
}e2s
2s s + 1 2(s + 2)
2s s + 1 2(s + 2)
Y (s) =
y(t) = L 1 {Y (s)}
1
1
1
1
= { e(t1) + e2(t1) }H(t 1) { e(t2) + e2(t2) }H(t 2)
2
2
2
2
3. (a) (6%) Find the inverse Laplace transform of F (s).

F (s) =
s2 + 2s + 3
(s2 + 2s + 2)(s2 + 2s + 5)
(b) (6%) Find the Laplace transform of f (t)

f (t) = eat
sinh t
t
( a is a constant)
100
(a)
L
s2 + 2s + 3
}
(s2 + 2s + 2)(s2 + 2s + 5)
(s + 1)2 + 2
1
}
L {
[(s + 1)2 + 1] [(s + 1)2 + 4]
s2 + 2
et L 1 { 2
}
(s + 1)(s2 + 4)
2
1
+
}
et L 1 { 2
2
3(s + 1) 3(s + 4)
1
1
et ( sin t + sin 2t)
3
3
{F (s)} = L 1 {
=
=
=
=
(b)
at sinh t
L {f (t)} = L {e
}=
s
1
ds
(s a)2 1
9934
1
ds
(s a 1)(s a + 1)
s

1
1
} ds
=
{
2(s a 1) 2(s a + 1)
s

1

{ln |s a 1| ln |s a + 1|}
=
2
s
sa+1
1
ln |
|
=
2
sa1
4. For the following ODE

x(x 1)y + (3x 1)y + y = 0
(a) (5%) Is x = 0 an ordinary point, regular singular point, or irregular singular
point ?
(b) (10%) Find power series solutions near x = 0.
100
(a) ODE
y +
P (x) =
3x 1
1
y +
y=0
x(x 1)
x(x 1)
1
3x 1
, Q(x) =
x(x 1)
x(x 1)
x = 0 P (x) Q(x) , x = 0 x P (x) =

x2 Q(x) =
x
, x = 0 ODE
x1
3x 1
x1
(b) x = 0 ODE ,
y=

n=0
an xn+r
(a0 = 0 , 0 < |x| < 1)
ODE ,
x2

n=0
an (n + r)(n + r 1)xn+r2 x
+3x

n=0
an (n + r)xn+r1

n=0

n=0
an (n + r)(n + r 1)xn+r2
an (n + r)xn+r1 +

n=0
an xn+r = 0
5.
9935
r 2 a0 xr1 +
[(n + r + 1)2 an (n + r + 1)2 an+1 ]xn+r = 0
n=0
r 2 a0 = 0
(n + r + 1)2 an (n + r + 1)2 an+1 = 0 n = 0 , 1 , 2
a0 = 0 , r2 = 0 , r = 0 , 0 ,
(n + r + 1)2 an (n + r + 1)2 an+1 = 0
an+1 = an ,
a1 = a0 , a2 = a1 = a0 , an = a0
y1 =

n+r
an x
n=0
r=0
= a0
xn = a0
n=0
1
1x

ln x
y

y2 =
= a0
xn+r ln x
= a0
xn ln x = a0

r r=0
r=0
1x
n=0
n=0
y(x) = c1 y1 + c2 y2 = c1
1
ln x
+ c2
1x
1x
5. (10%) Prove that let A be a real, symmetric matrix, then there is a real, orthogonal
100
matrix that diagonalize A.
Schurs , U ( U U 1 = U T ) , U 1 AU
,
d11
0
U 1 AU = U T AU =
d12
d22
d1n
d2n

dnn
A , AT = A, ,
d11
d
12
(U 1 AU)T = U T AU =

d1n

d11
0
0 0
=

0
d22
dnn
d12
d22
d1n
d2n

dnn
9936
dij = 0 ( i = j) , ,
d11
U 1 AU =
d22
dnn
6. If a vector eld F = 3x2 i + 2yz j + y 2 k , C : the path connecting (0 , 1 , 2) and

(1 , 1 , 7).
(a) (6%) Please show that F is a conservative vector eld. Since F is a conservative vector eld, there exists a potential function . Please also show the
relationship between F and
F d
r.
100
(b) (8%) Please calculate and
C
(a)

F =

x
3x2
y
2yz
F , , = F
z
y2

=0

(b) = F ,
= 3x2
3
x
= x + f1 (y , z)
= y 2z + f2 (x , z)
= 2yz
= y 2z + f3 (x , y)
=y
z
(x , y , z) = x3 + y 2z + c ,
(1 , 1 , 3)

F d
r = (x , y , z)
= (1 + 3) 2 = 2
C
(0 , 1 , 2)
5.
9937
7. (15%) Solving the following partial dierential equation (PDE) by separation of

variables.
2T
T
=
t
x2
I.C. T (x, 0) = f (x).
B.C. (1)
T (0, t)
=0
x
(2)
T ( , t)
= T ( , t).
x
100
(1) T (x , t) = X(x)Y (t)

T
2T
= X (x)Y (t) ,
= X(x)Y (t)
2
x
t
PDE
X(x)Y (t) = X (x)Y (t)
X
Y
=
= ( )
X
Y
X (x) + X(x) = 0
Y (t) + Y (t) = 0
(1)
(2) B.C.
T (0 , t)
= X (0)Y (t) = 0 X (0) = 0
x
T ( , t)
= T ( , t) X ( )Y (t) = X( )Y (t)
x
X ( ) = X( ) ,
X (x) + X(x) = 0 , X (0) = 0 , X ( ) = X( )
SturmLiouville
(i) < 0 , = p2 (0 < p < ) (2) ,
X (x) p2 X(x) = 0
(2)
9938
X(x) = c1 sinh px + c2 cosh px
X (x) = c1 p cosh px + c2 p sinh px
X (0) = 0, c1 = 0, X ( ) = X( ),
c2 p sinh p + c2 cosh p = 0
p sinh p + cosh p = 0, c2 = 0 , X(x) = 0 trivial

(ii) = 0 , (2) , X (x) = 0 ,
X(x) = c1 + c2 x X (x) = c2
X (0) = 0, c2 = 0, X ( ) = X( ), c1 = 0 ,
X(x) = 0 trivial
(iii) > 0 , = p2 (0 < p < ) (2) ,
X (x) + p2 X(x) = 0
X(x) = c1 sin px + c2 cos px
X (x) = c1 p cos px c2 p sin px
X (0) = 0, c1 = 0 , X ( ) = X( ),
c2 p sin p = c2 cos p
c2 (p sin p cos p ) = 0
c2 = 0,
p sin p cos p = 0
tan p =
1
p
(3) pn (n = 1 , 2 , 3 ) ,
X(x) = c2 cos pn x = Xn (n = 1 , 2 , 3 )
(3)
5.
9939
= p2 = p2n (1)
Y (t) + p2n Y (t) = 0
ODE,
2
Y (t) = depn t = Yn
Tn (x , t) = Yn (t)Xn (x) = depn t c2 cos pn x = An epn t cos pn x

(3)
T (x , t) =

n=1
Tn (x , t) =
T (x , 0) = f (x) =

n=1
An epn t cos pn x
An cos pn x
n=1
An =
< f (x) , cos pn x >

< cos pn x , cos pn x >
(4)
(4) An , An T (x , t)
8. You are given the following matrix of

0 4
A=
4 0
100
(a) (4%) Find the eigenvalues and eigenvectors of A.

(b) (4%) If A is similar to D (a diagonal matrix with eigenvalues as the diagonal
components), what are the transition matrix P and it inverse P 1 ?
(c) (4%) Use the results from (a) and (b) to solve the following system of ODEs :

y1 = 4y2 + 9t
y2 = 4y1 + 5
9940
(a) det(A I) = 2 + 16 = 0, A = 4i , = 4i
(A I)X = 0

4i
4
x1
0
=
4 4i
0
x2
X = c1

1
(c1 = 0)
= 4i (A I)X = 0

4i 4
4 4i

X = c2
(b)

P =

x1
x2

(c2 = 0)

P
(c)
y1
y2

AP = D =

0
=
0
0 4
=
4 0

1 i 1
=
2i i 1
4i
4i

9t
y1
+
5
y2
Y = AY + B

0 4
9t
y1
, A=
, B=
Y =
4 0
5
y2

z1
, (1)
Y = P Z, Z =
z2

P Z = AP Z + B Z = P 1 AP Z + P 1 B = DZ + P 1 B
z1
z2

4i
0
=
0 4i

z1
1 9t 5i
+
2 9t + 5i
z2
z1 = 4iz1 + (9t 5i)

2
z = 4iz + 1 (9t + 5i)

2
2
2
(1)
5.
9941
29
+
z1 =
32
z = 29
2
32
9t
i + c1 e4ti
8
9t
i + c2 e4ti
8
ODE

Y
29 9t
4ti
+ i + c1 e
1 1 32
y1
8
= PZ =
29
9t
y2
i i
i + c2 e4ti
32
8
29
+ c1 e4ti + c2 e4ti
16
9t
4ti
4ti
+ ic1 e ic2 e
4
29
+ cos 4t + sin 4t
16
( = c1 + c2 = (c1 c2 )i )
9t
sin 4t + cos 4t
4

9942
6. 100
1. Solve the following ordinary dierential equation (20%)
(1) y y 2y = 5 cos x
(2) A damped oscillation equation : my + cy + ky = 0 ; discuss the solution, if
(a) m = 1, c = 2, k = 2. (b) if m = 1, c = 4, k = 2.
100
(1)
(a) : y = emx ODE
m2 m 2 = 0
m = 2 1 ,
yh (x) = c1 e2x + c2 ex
(b) :
1
1
5 cos x = 5
cos x
D2
1 D 2
D3
D3
= 5
cos x = 5 2
cos x
(D + 3)(D 3)
D 9
1
1
( sin x 3 cos x) = (sin x + 3 cos x)
= 5
1 9
2
yp =
D2
(c) : y(x) = yh (x) + yp (x)

(2)
(a) m = 1 c = 2 k = 2 , ODE
y + 2y + 2y = 0
y(x) = emx ,
m2 + 2m + 2 = 0 m = 1 i
y(x) = ex (c1 cos x + c2 sin x)
6.
9943
(b) m = 1 c = 4 k = 2 ODE
y + 4y + 2y = 0
y(x) = emx ,
m2 + 4m + 2 = 0 m = 2
y(x) = c1 e(2+
2)x
+ c2 e(2
2)x
2. The following is the Lucas series (15%)

1 , 3 , 4 , 7 , 11 , 18 , 29 , 47 , 76 ,
a0 = 1 , a1 = 3 , a2 = 4 , a3 = 7 , a4 = 11 , a5 = 18 ,
(a) Please nd all mathematical rules.
an+1
=?
(b) Please nd lim
n an
1
= 1 + x + x2 + x3 + + xn +
(c) If F (x) =
1x
Please nd 1 1 + 1 1 + 1 1 + = ?
100
(a)
an + an+1 = an+2
a0 = 1 a1 = 3, an = M n , (1)
M n + M n+1 = M n+2
M n = 0,
1
2
M2 M 1 = 0
M = (1 5),
1+ 5 n
1 5 n
) + c2 (
)
an = c1 (
2
2
(1)
9944
a0 = 1 = c1 + c2
a1 = 3 = c1 ( 1 + 5 ) + c2 ( 1 5 )
2
2
1
1
c1 = (1 + 5) c2 = (1 5) ,
2
2
1 + 5 n+1
1 5 n+1
)
)
an = (
+(
(n = 0 , 1 , 2 , )
2
2
(b)
an+1
n an
lim
1 + 5 n+2
1 5 n+2
(
+(
)
)
2
2
= lim
n 1 +
1 5 n+1
5 n+1
(
+(
)
)
2
2
1 5 1 5 n+1
1+ 5
)
)+(
)(
(
2
2
1
+
5
= lim
n
1 5 n+1
)
1+(
1+ 5
1+ 5
=
2
(c)
11+11+ =
(1)n 1 =
n=0
(1)n an
n=0
lim an = lim 1 = 1 = 0
11+11+ =
(1)n 1
n=0
2 1 0 0
0 2 1 0
3. Given A =
0 0 2 1
0 0 0 2
(a) nd eigenvalues and eigenvectors of A. (10%)
(b) nd all solution of (A 2I)X = (1 0 0 0)T . (10%)
(c) nd eigenvalues and eigenvectors of (A 2I)100 . (10%)
100
6.
9945
(a) det(A I) = 0, = 2 2 2 2 , = 2 (A I)X = 0,

0
0 0 1 0x 0
=
0 0 0 1 x3 0
0
x4
0 0 0 0
0 1 0 0

1
0

X = c1
0
x1
(c1 = 0)
0
(b) (A 2I)X = (1 0 0 0)T ,

1
0 0 1 0x 0
=
0 0 0 1 x3 0
0
x4
0 0 0 0
0 1 0 0
x1
x2 = 1 x3 = 0 x4 = 0 x1 = c,

1
0
x1
0 1
x

2
X = = c +
0 0
x3
x4
(c)
0 1 0 0
0 0 1 0
B = (A 2I) =
0 0 0 1
0 0 0 0
det(B I) = 0, B = 0 0 0 0 ,

1
0

X = c1
0
(c1 = 0)
B 100 = (A 2I)100 0 0 0 0 , X
9946
4. Consider the boundary-value problem y + y = 0, y(0) = y(2), y (0) = y (2).

Show that except for the case = 0, there are two independent eigenfunctions
corresponding to each eigenvalue. (15%)
100
(a) < 0 ; = p2 (0 < p < ) , ODE

y (x) p2 y(x) = 0
ODE
y(x) = c1 epx + c2 epx
(1)
y (x) = p c1 epx p c2 epx
(2)
y(0) = y(2) (1) ,

c1 + c2 = c1 e2p + c2 e2p
c1 (1 e2p ) + c2 (1 e2p ) = 0
(3)
y (0) = y (2) (2) ,

p c1 p c2 = p c1 e2p p c2 e2p
c1 (1 e2p ) c2 (1 e2p ) = 0
(4)
(3) (4) , c1 c2 ,

1 e2p

1 e2p

1 e2p
= 0
1 + e2p
Cramers c1 = c2 = 0 , c1 = c2 = 0 (1) , y(x) = 0 ,

ODE ( trivial solution)
(b) = 0 , y (x) = 0 , ODE
y(x) = c1 x + c2 y (x) = c1
(5)
6.
9947
y(0) = y(2) (5) ,

c2 = c1 2 + c2
c1 = 0 , y (0) = y (2) (5) , c1 = c1 , ,

c2 , c2 = 0 , = 0 ODE ,
y(x) = c2 (c2 = 0)
(c) > 0 ; = p2 (0 < p < ) , ODE
y (x) + p2 y(x) = 0
ODE
y(x) = c1 cos px + c2 sin px
(6)
y (x) = p c1 sin px + p c2 cos px
(7)
y(0) = y(2) (6) ,

c1 = c1 cos 2p + c2 sin 2p
c1 (1 cos 2p) c2 sin 2p = 0
(8)
y (0) = y (2) (7) ,

p c2 = p c1 sin 2p + p c2 cos 2p
c1 sin 2p + c2 (1 cos 2p) = 0
Cramers , (8) (9) c1 c2 0

1 cos 2p

sin 2p

sin 2p
=0
1 cos 2p
(1 cos 2p)2 + sin2 2p = 0
1 2 cos 2p + cos2 2p + sin2 2p = 0 2 2 cos 2p = 0
cos 2p = 1 , 2p = 2n ,
p = n (n = 1 , 2 , 3 )
(9)
9948
ODE
= p2 = n2 = n2
(n = 1 , 2 , 3 )
y(x) = c1 cos nx + c2 sin nx (c1 c2 0)
5. Find the Taylor series of the following functions about the point z = a and determine the radius of convergence. (20%)
(1) ln z, a = 1.
(2) cos z 2 , a = 0.
100
(1) t = z 1, z = 1 , t = 0 ,
ln z = ln(1 + t) =
(1)n
(1)n

tn+1 =
(z 1)n+1
n=0 n + 1
n=0 n + 1
|z 1| < 1, 1
(2) cos z =
(1)n
(1)n

z 2n , cos z 2 =
z 4n , |z| < ,
(2n)!
(2n)!
n=0
n=0
7. (C)
9949
7. 100 (C)
1 13 (, , ) ,
, 14 15 ,
1. (10%) Here is an augmented matrix in which denotes an arbitrary number and

# denotes a nonzero number.
#
0 #
0 0 #
0
0 #
Which of the following statements is (are) true ?

(A) The given augmented matrix is consistent and the solution is unique.
(B) The given augmented matrix is inconsistent and the solution is unique.
(C) The given augmented matrix is consistent and the solution is not unique.
(D) The given augmented matrix is inconsistent and the solution is not unique.
100 C
(E) None of the above.
A 4 5 B 4 1 ,
#
0 #
[A|B] =
0 0 #
0 0 0

0 #
rank (A) = rank [A|B] = 4 , (C)
2. (5%) Assume A is an m n matrix . If rank A = n, then

(A) The columns of A are linearly independent.
(B) Ax = b has at least one solution for every b in Rm .
(C) Every column of its reduced row echelon form contains a pivot position.
(D) Ax = b has at most one solution for every b in Rm .
(E) None of the above statements is true.
100 C
9950
(A) True ; rank (Amn ) = n , A

(B) False ; rank (A) = rank [A|b] Ax = b
(C) True ;
(D) True ; rank (A) = rank [A|b] = n , Ax = b ,
rank (A) = n = rank [A|b] , Ax = b , Ax = b
(A)(C)(D)
3. (5%) For the four vectors below , which of the following statements is (are) true ?

3
8
1
2

2
0
2
1
0
6
3
6
(A) They are linearly independent.
(B) They span R3
(C) They are linearly dependent.
(D) They do not span R3
100 C
3 8 1 2
rank 2 0 2 1 = 3 , (B)(C)
0 6 3 6
4. (5%) Let V be a subspace of Rn with dimension k. Which of the following statements is (are) true ?
(A) Every linearly independent subset of V contains at least k vectors.
(B) Any nite subset of V containing more than k vectors is linearly dependent.
(C) n k .
(D) Any nite subset of V containing less thank k vectors is linearly independent.
100 C
7. (C)
9951
(A) False ; at most k vectors

(B) True ;
(C) True ; dim (Rn ) dim (V )
(D) False ;
(B)(C)
5. (5%) Suppose that s, t, and u are vectors in Rn such that s is orthogonal to u and
u is orthogonal to t. Then
(A) s is orthogonal to t
(B) For any orthogonal n n matrix P , we have that P u is orthogonal to both s
and t
(C) For any orthogonal n n matrix P , we have that P s is orthogonal to u
(D) s + t is orthogonal to u
(E) None of the preceding statements are true.
100 C

1
0
2
(A) False ; , s =
,u=
,t=
0
1
0
(B) False ;
(C) False ;
(D) True ; < s , u >= 0 < t , u >= 0, < s + t , u >=< s , u > + < t , u >= 0
(D)
9952
6. (5%) Let M be an inner product space

(A) Since the inner product is an integral of a product of functions, M is a function
space
(B) c < u , v >=< cu , cv > for all vectors u and v in M and for every scalar c
(C) < T (u) , v >=< u , T (v) > for all vectors u and v in M and for every linear
operator T on M
(D) If M is nite-dimensional, then M contains an orthonormal basis
(E) None of the preceding statements are true.
100 C
(A) True ;
(B) False ; < cu , cv >= c
c < u, v >
(C) False ; < T (u) , v >=< u , T (v) >
(D) True ;
(A)(D)
7. (5%) An m n matrix P is invertible if

(A) The reduced row echelon form of P is In
(B) The column of P are linearly independent
(C) The column of P span Rm
(D) The rows of P are linearly independent
(E) None of the preceding statements is true.
P n n, m = n , (A)(B)(C)(D)
100 C
7. (C)
9953
8. (5%) Suppose that S is an arbitrary n n matrix. Then

(A) det(S) is the sum of its diagonal entries
(B) det S 2 = 2 det S
(C) det S is a vector in Rn
(D) det S = det S T
100 C
(A) False ;
(B) False ; det(S 2 ) = (det(S))2
(C) False ; det(S) R
(D) True ;
(D)
9. (5%) Let A be a subset of R3 containing two or more vectors. Then :

(A) The span of any two vectors in A is a plane in R3 .
(B) If A contains more than three vectors, then A is linearly independent
(C) The span of any two nonzero vectors in A is a plane in R3 .
(D) Every vector in A is in the span of A.
(A) False ; , 2
(B) False ; A 3 ,
(C) False ; (A)
(D) True ;
100 C
9954
(D)
10. (8%) The solution for f (t) of the following equation
t
2
t
f ()et d
f (t) = 3t e
0
Has the following form : f (t) = g(t) k1 ek2 t , where k1 , k2 are constants and g(t)
is a function containing only polynomial terms like kn tn . Which of the following
item(s) is (are) true :
(A) k1 + k2 = 3 (B) k2 + g(0) = 1 (C) g(1) + g(0) = 4 (D) g(1) + g(1) = 8
100 C
(E) none of above.
f (t) = 3t2 et f (t) et
(1)
(1) LT ,
F (s) =
1
6
1
F (s)
3
s
s+1 s1
L {f (t)} = F (s) ,
F (s) =
6
2
(s 1)(s3 + 6s + 6)
6
1
= 3 4+
4
s (s + 1)
s
s
s s+1
f (t) = L {F (s)} = 3t2 t3 + 1 2et
g(t) = 3t2 t3 + 1 k1 = 2 k2 = 1 , (C)(D)
11. (7%) The function f (x) = |x| can be expressed as a Fourier series on the interval

x , where f (x) = |x| = a0 +
[an cos(nx) + bn sin(nx)]
n=1
Which of the following item(s) is (are) true :

(A) a0 =
(B) a2 + b3 = 0
(E) none of above .
(C) b2 + b4 = 0
(D) a1 + a3 =
4
9
100 C
7. (C)
9955
f (x) = |x| ,
1
a0 =
2
an =
x dx =
0
x cos(nx) dx =
0
2[(1)n 1]
n2
bn = 0 , (B)(C)
12 13 We are going to solve of the following dierential equation system :
dx1
= 3x1 + x2 x3
dt
dx2
= x1 + 3x2 x3
dt
dx3 = 3x1 + 3x2 x3

dt
with initial values as x1 (0) = 4, x2 (0) = 7 and x3 (0) = 7. The solution has the
form as
1 t
2 t
3 t
x1 = A11 e + A12 e + A13 e

x2 = A21 e1 t + A22 e2 t + A23 e3 t
x3 = A31 e1 t + A32 e2 t + A33 e3 t
where |1 | |2 | |3 | and Aij are constants. Which of the following item(s) is

(are) true :
(12.) (5%) (A) 1 + 2 = 3 (B) 21 + 3 = 3 (C) 2 + 3 = 3 (D) 1 + 3 = 3
(E) none of above.
(13.) (5%) (A) x1 (1) + 2x2 (1) = 6e (B) x1 (1) + x3 (1) = 2 cos(1)
(C) x2 (1) + x3 (1) = 8e
(D) x1 (1) x2 (1) = 2 cos(1) (E) none of above.

100 C

3 1 1
x1

A = 1 3 1 , X = x2
x3
3 3 1
9956
det(A I) = 0 , = 1 2 2 , = 1 (A I)V = 0

1 2 1 v2 = 0
v3
3 3 2
0
2 1 1
v1

1

0) V1 = 1
(c1 =

1

V = c1 1
= 2 (A I)V = 0

1 1 1 v2 = 0
v3
3 3 3
0
1 1 1
1
1
V = c2 1 + c3 0
v1
( c2 c3 0)
V2 = 1 , V3 = 0
0
1
1
X = 1 V1 et + 2 V2 e2t + 3 V3 e2t

1
1
1
t
2t
2t
= 1 1 e + 2 1 e + 3 0 e
3
0
1
x1 (0)
1
1
1
4
X(0) = x2 (0) = 7 = 1 1 + 2 1 + 3 0
x3 (0)
1 = 4 2 = 11 3 = 19 ,
1
1
4et 11e2t + 19e2t
x1 (t)
1
2t
t
2t
t
2t
2t
x2 (t) = 4 1 e 11 1 e + 19 0 e = 4e + 11e + 0e
12et + 0e2t + 19e2t
x3 (t)
0
1
3
7. (C)
9957
12 (A)(D) , 13 (E)
14. (19%) The initial value problem

1 d2 x 6 dx
1
+
+ 2x = 5 cos(4t) , x(0) = , x (0) = 0
2
5 dt
5 dt
2
has the solution with the form :
x(t) =
1
{f (t)[k1 cos(t) + k2 sin(t)] + k3 cos(k4 t) + k5 sin(k4 t)}
102
k1 , k2 , k3 , k4 , k5 are constants. f (t) is a function of t . Find k1 , k2 , k3 , k4 , k5 and

f (t).
100 C
(1) : x = emt
dx
d2 x
+ 6 + 10x = 25 cos(4t)
2
dt
dt
(1)
m2 + 6m + 10 = 0 m = 3 i
xh = e3t (c1 cos t + c2 sin t)

(2) :
1
1
25 cos(4t) =
25 cos(4t)
+ 6D + 10
16 + 6D + 10
(D + 1)
25 D + 1
25
cos(4t) =
cos(4t)
=
6 (D 1)(D + 1)
6 D2 1
1
1
25
(4 sin 4t + cos 4t) =
(100 sin 4t 25 cos 4t)
=
6 16 1
102
xp =
D2
(3) :
x(t) = xh (t) + xp (t) = e3t (c1 cos t + c2 sin t) +
1
(100 sin 4t 25 cos 4t)
102
1
25
x(0) = = c1
2
102
x (0) = 0 = c 3c + 400
2
1
102
(1)
9958
c1 =
38
86
c2 = ,
51
51
x(t) = e3t (
86
1
38
cos t
sin t) +
(100 sin 4t 25 cos 4t)
51
51
102
15. (6%) The dierential equation

x2
d2 y
dy
x + y = ln(x)
2
dx
dx
has the solution with the form :

y(x) = c1 x + c2 g1 (x) + p1 + g2 (x)
c1 and c2 are arbitrary constants P1 is a constant. g1 (x) and g2 (x) are function of
100 C
x. Find p1 and g1 (x)
x = et , t = ln x,
xy = Dy , x2 y = D(D 1)y , x3 y = D(D 1)(D 2)y
D =
d
, ODE [ D
dt
D(D 1)y Dy + y = t
(D 1)2 y = t
(1) : y = emt (1)
(m 1)2 = 0 m = 1 , 1
yh = c1 et + c2 tet = c1 x + c2 x ln x
(2) :
yp =
1
1
t=
t = (1 + 2D + )t = t + 2 = 2 + ln x
2
(D 1)
1 2D + D2
(3) : y(x) = yh + yp = c1 x + c2 x ln x + 2 + ln x , p1 = 2, g1 (x) = x ln x
(1)
8.
9959
8. 100

2 4
1. (20%) Give the following matrix A =
3 1
(a) (5%) Find the eigenvalues of A.
(b) (5%) Find the eigenvectors of A.

(c) (5%) Diagonalize A, i.e., nd matrices P and D such that P 1AP = D, where
D is a diagonal matrix.
(d) (5%) Find A6 ( By using the results from (c)).
100
(a)
det(A I) = (1)2 (2 3 10) = 0
= 5 2
(b) = 5 (A I)X = 0

3 4
x1
3
X = c1
x2

4

0
0
(c1 = 0)
= 2 (A I)X = 0

4 4
3 3

X = c2
(c)
x1
x2
1
1

0
=
0

(c2 = 0)
4 1
P =
3 1
5 0
AP = D =
0 3
9960

6
(d) A = P D P
=P
(3)6
2. (15%) Let L {f (t)} =
56
f (t)est dt be the Laplace transformation of f (t).
1
(a) (5%) Show that L {t} = 2 .
s
(b) (5%) Show that L {f (t)} = s2 L {f (t)} sf (0) f (0).
(c) (5%) Find L {cos(wt + )}, where w and are constants.
100
(a)
L {t} =
st
te
0
1 st
1 st
1
t
e
dt = (
2 e ) = 2
s
s
s
0
(b) f (t) f (t) f (t) t 0 ,

f (t)est dt
L {f (t)} =
0
st
= {e
st
f (t) + se

2
f (t)} + s
0
= f (0) sf (0) + s2 L {f (t)}
f (t)est dt
(c)
L {cos(wt + )} = L {cos wt cos sin wt sin }
s cos
w sin
= 2
2
2
s +w
s + w2
3. (15%) Given a vector F = y 2 i + x2 j + z 2 k , and a cylindrical surface S of

x2 + y 2 4, 0 z 5. Using the Gauss divergence theorem, evaluate the surface
integral.
100
S x2 + y 2 4 , 0 z 5 , D ,
(F
n ) dA =
( F ) dxdydz
D
8.
9961
2 z dxdydz = 2 z V = 2
=
D
5
22 5
2
= 100
100
4. (a) (10%) Find the general solution of
dy(x)
= x3 (y x)2 + x1 y , x > 0
dx
(b) (5%) If we further require y(1) = 2, what is the solution of y(x).
(a) u = (y x) ,
du
dy
=
1, ODE
dx
dx
1
du
+ 1 = x3 u2 + (u + x)
dx
x
du 1
u = x3 u2
dx x
u2 ,
u2
v=
u

I = e
1
x
dx
du 1 1
u = x3
dx x
dv 1
+ v = x3
dx x
= eln |x| = x,
1
Iv = x(x3 ) dx = x5 + c
5
v=
1
1
c
= x4 +
u
5
x
ODE
1
c
1
= x4 +
yx
5
x
1
2
1
(b) x = 1 y = 2 (1)
= + c, c = ,
3
5
15
1
2
1
= x4
yx
5
15x
(1)
9962
5. (a) (5%) Find the Fourier series expansion for the following function
f (x) = x + x2 , < x <
(b) (3%) Express the Parsevals relation in terms of the Fourier coecients a0 , ak ,
bk , k = 1 , 2 , .
(c) (7%) Can you prove that the Fourier coecients in (a) satisfy the Parsevals
1
1

4
2
=
=
and
)
100
relation. (Hint :
4
2
90
6
k=1 k
k=1 k
(a)
f (x) = a0 +
(ak cos kx + bk sin kx)
k=1
1
(b)
2

1
2
a0 =
(x + x2 ) dx =
2
3

1
4 cos k
(x + x2 ) cos kx dx =
ak =

k2
1
2 cos k
bk =
(x + x2 ) sin kx dx =

k
f 2 (x) dx = a20 +
(c)
1
2
a2

b2
( k + k)
2
k=1 2
1
f (x) dx =
2
a20 = (
(x2 + x)2 dx =
2 4
+
3
5
2 2 4
) =
3
9
a2
1 4 cos k
1

b2
2 cos k 2
2
(
(
)
( k + k) =
)
+
2
2
k
k
k=1 2
k=1 2
k=1 2
1
1

= 8
+
2
4
2
k=1 k
k=1 k
2
4 4 2
4
+2
=
+
= 8
90
6
45
3
(1)
(2)
(3)
8.
9963
(1) (2) (3) ,

a20
a2

b2
4 4 4 2
4 2
1
+
+
=
+
=
+
( k + k) =
2
9
45
3
5
3
2
k=1 2
f 2 (x) dx
6. (20%) For the one-dimensional heat conduction equation :

ut = uxx + f (u) , x R , t > 0
(1)
the solution with u(x , t) = q(x ct) is called a wave solution, where c R is the
dq
d2 q
wave propagation speed. Let = x ct, and q =
and q = 2 .
d
d
(a) (5%) Derive an ordinary dierential equation for q(), that Eq.(1) has a wave
solution u = q(x ct).
(b) (15%) Let
f (u) = u(1 u)(u 0.3)
Derive the wave solution u(x ct) of Eq.(1) with the above f (u). ( Hint
q = kq(q 1), and c and k are to be determined ). Write k = ? and c = ?
100
(a) = x ct,
du
du
u
=
= c
t
d t
d
du
du 2 u
d2 u
u
u
=
=
,
(
)
=
=
x
d x
d x2
x x
d 2
PDE
c
(b) hint
d2 u
du
= 2 + f (u)
d
d
d2 u
du
+ f (u) = 0
+
c
d 2
d
du
= ku (u 1)
d
du
1
1
= k d (
)du = k d
u(u 1)
u1 u
(1)
9964
ln |u 1| ln |u| = k +
u1
= ek
u
u=
(2)
( )
1
1 ek
(2)
d2 u
du
+ c + u(1 u)(u 0.3) = 0
2
d
d
ek (0.7 c k k 2 + ek (0.3 + c k k 2 ))
=0
(1 + ek )3
(0.7 c k k 2 ) + ek (0.3 + c k k 2 ) = 0
0.7 c k k 2 = 0
0.3 + c k k 2 = 0
c = 0.282843
k = 0.707107
(3)
u=
c = 0.282843
k = 0.707107
1
1
=
k
1 e
1 ek (xct)
(3)
9.
9965
9. 100
1. (15%) Which of the following is (are) correct ? ()
(a) T r(AB) = T r(BA) holds for arbitrary nite or innite dimensional matrices
A, B.
(b) For all 33 matrices A, it is possible to express det(A) as a function of T r(A),
T r(A2 ), T r(A3 ).
(c) The 3 functions 1, x, x2 are linear independent.
(d) The matrix U = exp(A) is unitary whenever A is anti-Hermitian.
(e) Arbitrary vectors A,B, C in 3 dimensional Euclidean space satisfy the relation
A (B C) = B(A C) + C(B A), where denotes the cross product,
and denotes the inner product.
100
(a) False ;
(b) False ; A 1 2 3
1 2 3 = (31 + 32 + 33 ) + (21 + 22 + 23 ) + (1 + 2 + 3 )

det(A) = T r(A3 ) + T r(A2 ) + T r(A)

1 x x2

(c) True ; W (1 , x , x2 ) = 0 1 2x = 2 = 0 , {1 , x , x2 }

0 0 2
(d) True ;
UU = (exp(A))(exp(A)) = (exp(A))(exp(A )) = exp(A + A ) = I
A + A = 0 , A = A , A
(e) False ; A (B C) = B(A C) C(B A)
9966
2. (a) (15%) Find the eigenvalues and corresponding eigenvectors of the matrix

3/5 4/5
M=
4/5 3/5
(b) (10%) Find the value of T r(f (M)) for f (x) = x2 + sin x.
100
(a)
det(A I) = (1)2 (2 1) = 0
= 1, = 1 (A I)X = 0

2/5 4/5
4/5 8/5
X = c1

2
8/5 4/5

4/5 2/5

X = c2
1
2
x1
x2

0
=
0
(c1 = 0)
= 1 (A I)X = 0

x1
x2

=

0
0

(c2 = 0)
(b) f (M) = M 2 + sin M 1 + sin 1 (1)2 + sin(1),

T r(f (M)) = (1 + sin 1) + [(1)2 + sin(1)] = 2
3. (15%) Find the general solution for y(x) to the dierential equation
d2 y
dy
+ y = exp(x)
2
dx2
dx
100
9.
9967
(1) : y(x) = emx ODE

m2 2m + 1 = 0 m = 1 , 1
yh (x) = c1 ex + c2 x ex
(2) :
yp (x) =
1
x2 ex
x
=
e
D 2 2D + 1
2
(3) : y(x) = yh (x) + yp (x)
4. (15%) Solve the dierential equation

dy
y2
=
dx
1 + x2
100
for the solution with y(0) = 1.
dy
1
=
dx
2
y
1 + x2
1
1
= tan1 x + c y(x) =
1
y
tan x + c
1
c
y(0) = 1 = , c = 1,
y(x) =
1
tan x 1
1
5. (15%) For all non-negative integers n, evaluate the integral

xn exp(x2 /2) dx
100
9968
(1) n = 2m + 1 (m = 1 , 2 , 3 )
n
2
x exp(x /2) dx =
x2m+1 exp(x2 /2) dx = 0
(2) n = 2m
n
2
x exp(x /2) dx =
x2m exp(x2 /2) dx
dv = xex
u = x2m1
2 /2
dx
x2 /2
dv = (2m 1)x2m2 dx
v = e

2
2

= x2m1 ex /2
+ (2m 1)
x2m2 ex /2 dx
= (2m 1)(2m 3) 3 1
= (2m 1)(2m 3) 3 1
(2n)!
2
=
2n n!
6. (15%) Dene F (k) by the equality

F (k) exp(ikx) dk =
2 /2
dx
100
f (x) =
1
,
1 + x2
1
F (k) =
2
g(z) =
ex
1
1 + x2
Find F (k).
1
eikx dx
2
1+x
eikz
, g(z) z = i pole,
1 + z2
Resf (i) =
(1) k > 0
1
F (k) =
2
ek
ek
, Resf (i) =
2i
2i
1
ek
ek
1
ikx
(2
i)
Resf
(i)
=
i
=
e
dx
=
1 + x2
2
2i
2
(1)
9.
9969
(2) k = 0
1
F (k) =
2

1
1
1
1
tan x
dx =
=
2
1+x
2
2
(3) k < 0
1
F (k) =
2
(4) : F (k) =
1
ek
ek
1
ikx
2
i
Resf
(i)
=
i
=
e
dx
=
1 + x2
2
2i
2
e|k|
( < k < )
2
9970
10. 100
1. Find the general solution of the dierential equation (20%)
y = (2x + y)2 7
100
u = (2x + y),
du
dy
= 2 +
, ODE
dx
dx
du
du
+ 2 = u2 7
= u2 9
dx
dx
du
= dx
9
u2
1
1
{
} du =
6(u 3) 6(u + 3)
dx
1
u3
ln |
| = x + c1
6
u+3
, ODE
2x + y 3
= c e6x
2x + y + 3
2. Solve the initial value problem (20%)

y 4y + 4y = f (t) ; y(0) = y (0) = 0 with f (t) =
for 0 t < 3
t + 2 for t 3
100
f (t) = t{H(t) H(t 3)} + (t + 2) H(t 3)
10.
9971
L {f (t)} = L {t + 2H(t 3)} =
1
2e3s
+
s2
s
ODE LT
s2 Y (s) sy(0) y (0) 4{sY (s) y(0)} + 4Y (s) = L {f (t)} =
1
2e3s
+
s2
s
L {y(t)} = Y (s),
(s2 4s + 4)Y (s) =
1
2e3s
+
s2
s
1
2e3s
+
s2 (s 2)2 s(s 2)2
1
1
1
1
1
1
1
=
+ 2+
+{
+ }e3s
2
2
4(s 2)
4(s 2) 4s
4s
(s 2)
2(s 2) 2s
Y (s) =
y(t) = L 1 {Y (s)}
1 2t 1 2t t 1
1
1
te e + + + {(t 3)e2(t3) e2(t3) + }H(t 3)
=
4
4
4 4
2
2
3. Use an orthogonal transformation to transform the real quadratic form

Q = 4x1 x2 + 4x2 x3 + 4x3 x1
into the canonical form. Write down the canonical form and the transformation
matrix. (20%)
100
Q = X T AX,

0 2 2
x1

A = 2 0 2 , X = x2
2 2 0
x3
det(A I) = (1)3 (3 12 16) = 0
9972
A = 4 2 2 , = 4 (A I)V = 0

4 2
2
0
v1

2 4 2 v2 = 0
v3
2
2 4
0

1

V = c1 1

1

0) V1 = 1
(c1 =
1
1
= 2 (A I)V = 0

2 2 2
0
v1

2 2 2 v2 = 0
v3
2 2 2
0
1
1
V = c2 1 + c3 1
0
2
( c2 c3 0 )
V2 = 1 , V3 = 1
0
2
1
1
1
2
6
3

V
1
V2
V3
1
1
1

=
S=
3
||V1 || ||V2 || ||V3 ||
2
6
1
2
3
6
T
1
S , S = S , X = SY , Y = [y1 y2 y3 ]T ,
Q = X T AX = (SY )T ASY = Y T S T ASY = Y T S 1 ASY

4 0
0
y1

T
= Y DY = [y1 y2 y3 ] 0 2 0 y2
y3
0 0 2
= 4y12 2y22 2y32
10.
9973
4. Verify Greens theorem for the case where the vector eld
F (x , y) = (xy + y 2) i + x2 j
and the region is bounded by the curves y = x and y = x2 .
(Note : Calculate each side of Greens theorem, respectively, and show that both
100
results are identical.) (20%)
(1) : C = C1 + C2 () , C1 : y = x2 , x = 0 1 C2 : y = x ,
x=10,

F dr =
(xy + y 2 ) dx + x2 dy
C
C
2
2
(xy + y ) dx + x dy +
(xy + y 2 ) dx + x2 dy
=
C1
C2
x=0
x=1
2
2 2
2
2
{[x(x ) + (x ) ] dx + x d(x )} +
{(x x + x2 )dx + x2 dx}
=
x=0
x=1
x=1
x=0
3
4
3
=
(x + x + 2x )dx +
3x3 dx
x=0
x=1
19
1
=
1 =
20
20
(2) : C D ,
x2
(xy + y 2 )} dxdy =
{
x
y
D
x=1
y=x
x=0
1
20
y=x2
(2x x 2y) dydx
9974
(3) (1) (2)

x2
2
2
(xy + y 2 )} dxdy
F dr = (xy + y ) dx + x dy =
{
x
y
C
C
D
Greens
5. Starting from the separation of variables, solve the boundary-value problem

1 2w
2 w 1 w
+
+
=0; 1r2, 0
r 2
r r
r 2 2
with w(r , 0) = w(r , ) = w(1 , ) = 0, and w(2 , ) = sin3 . (20%)
100
(1) w(r , ) = R(r)()

w
2w
2w

= R (r)() ,
=
R
(r)()
,
= R(r) ()
r
r 2
2
PDE
()
r 2 R (r) + rR (r)
=
= ( )
R(r)
()

() + () = 0
r 2 R (r) + rR (r) R(r) = 0
(1)
(2) w(r , ) = R(r)() , B.C.

w(r , 0) = R(r)(0) = 0 (0) = 0
w(r , ) = R(r)() = 0 () = 0
() + () = 0 , (0) = () = 0
SturmLiouville
(i) < 0 = p2 (0 < p < ) , (2)
() p2 () = 0
(2)
10.
9975
() = c1 sinh p + c2 cosh p
(0) = 0 , c2 = 0 , () = 0 = c1 sinh p , c1 = 0 ,
() = 0 , w(r , ) = R(r)() = 0
(ii) = 0 (2) , () = 0 ,
() = c1 + c2
(0) = 0 , c1 = 0 , () = c2 , c2 = 0, () = 0,
w(r , ) = R(r)() = 0
(iii) > 0 = p2 (0 < p < ) , (2)
() + p2 () = 0
() = c1 sin p + c2 cos p
(0) = 0 , c2 = 0 , () = 0 = c1 sin p , c1 = 0 ,
sin p = 0 ,
p = n (n = 1 , 2 , 3 , )
(n = 1 , 2 , 3 , )
() = c1 sin n = n
= p2 = n2 (1)
r 2 R (r) + rR (r) n2 R(r) = 0
Euler - Cauchy ODE ,

R(r) = d1 r n + d2 r n = Rn
wn (r , ) = Rn n = (d1 r n + d2 r n ) c1 sin n = (An r n + Bn r n ) sin n

(3)
w(r , ) =

n=1
wn (r , ) =
w(1 , ) = 0 =
(An r n + Bn r n ) sin n
n=1
(An + Bn ) sin n
n=1
9976
An + Bn = 0
w(2 , ) = sin3 =
(1)

1
3
sin sin 3 =
(An 2n + Bn 2n ) sin n
4
4
n=1
A1 2 + B1 21 =
1
A3 23 + B3 23 =
n
n
An 2 + Bn 2 = 0 (n = 1 , 3)
(1) (2)
A1 =
1
1
2
2
, B1 = , A3 =
, B3 =
2
2
63
63
An = Bn = 0,
1
2r 3
2
r
+ 3 ) sin 3
w(r , ) = ( ) sin + (
2 2r
63
r 63
(2)
11.
9977
11. 100
1. : partial dierential equation ? Linear dierential equation ?
100
Solution of dierential equation ? (12%)
(1) partial dierential equation : ,

(2) Linear dierential equation :
(a)
(b)
(c)

(3) Solution of dierential equation : ODE , ODE
2. Find the solution of DE : y 3y + 2y = cos(ex ).
(20 %)
100
(1) : y = emx ODE

m2 3m + 2 = 0 m = 1 , 2
yh (x) = c1 ex + c2 e2x
(2) :
yp (x) =
1
cos(ex )
(D 1)(D 2)
9978
= e
2x
ex cos(x) dx
(12)x
2x
= e
ex { sin(ex )} dx
= e2x { cos(ex }
(3) : y(x) = yh (x) + yp (x) = c1 ex + c2 e2x e2x cos(ex )
3. Solve the initial value problem : y 4y + 13y = 4(t 2), y(0) = 0, y (0) = 1.
(16%)
100
ODE LT
s2 Y (s) sy(0) y (0) 4{sY (s) y(0)} + 13Y (s) = 4e2s
L {y(t)} = Y (s),
(s2 4s + 13)Y (s) = 1 + 4e2s
1 + 4e2s
1 + 4e2s
=
Y (s) = 2
s 4s + 13
(s 2)2 + 32
1
4
y(t) = L 1 {Y (s)} = e2t sin 3t + e2(t2) sin 3(t 2) H(t 2)
3
3
4. Evaluate the line integral with respect to arc length :

parabola: z = y 2, x = 1 for 0 y 2. (12%)
y=2
yz dz =
C
y=2
y(y )d(y ) =
y=0
y=0
2y 4 dy =
yz dz, with C the
100
2y 5 2 64
=
5 0
5
11.
9979
3 0 2
5. For systems of linear dierential equations, X = AX, A = 0 2 0 .

2 0 0
Please nd out if A is diagonalizable(show clearly the processes), also nd the
fundamental matrix, and the general solution. (16%) 100
det(A I) = ( 2)( 3 4) = 0
= 2 1 4 , = 2 (A I)V = 0
1 0 2
v1
0 0 0 v2 = 0
v3
2 0 2
0

V = c1 1
0

(c1 = 0) V1 = 1
0
= 1 (A I)V = 0
4 0 2
0
v1
0 3 0 v2 = 0
v3
2 0 1
0

V = c2 0
2

(c2 = 0) V2 = 0
2
= 4 (A I)V = 0
1 0 2
0
v1
0 2 0 v2 = 0
v3
2 0 4
0
9980
V = c3 0
1
(c3 = 0) V3 = 0
1
S = [V1 V2
0 1 2
V3 ] = 1 0 0
0 2 1
S 1 AS = D = 0 1 0
0 0 4
X = SY , Y = [y1 y2 y3 ]T , X = AX
SY = ASY Y = S 1 ASY = DY

2t
y1 = e
y1 = 2y1
y2 = y2
y2 = et

y3 = 4y3
y3 = e4t
ODE
X = SY = [V1 V2
e2t
V3 ] et
e4t
= V1 e2t + V2 et + V3 e4t
t
0
2e4t
e
= e2t + 0 + 0
0
2 et
e4t
fundamental matrix
0 et 2e4t
2t
0
0
e
0 2 et e4t
11.
9981
6. For a thin, homogeneous bar of length L, given the initial temperature throughout
the bar is f (x), the temperature at both ends at all time are zero. i.e.
2
= a2 2 ; 0 < x < L , t > 0

t
x
(0 , t) = (L , t) = 0, and (x , 0) = f (x). Please determine the temperature
100
distribution (x , t) on the bar. (16%)
nx
,
L n=1

nx
(x , t) =
an (t) sin
L
n=1
(0 , t) = (L , t) = 0,
sin
PDE

n=1
an (t) sin

nx
n2 2
nx
= a2
an (t)( 2 ) sin
L
L
L
n=1

n=1
an (t) + (

na 2
nx
) an (t) sin
=0
L
L
an (t) + (
an (t) = An e(
na 2
) t
L
na 2
) an (t) = 0
L
, PDE
(x , t) =
An e(
sin
nx
L
An sin
nx
L
na 2
) t
L
n=1
I.C.
(x , 0) = f (x) =

n=1
2
An =
L
f (x) sin
0
nx
dx
L
An (1) ,
7. Please nd all z such that ez = 2 + i. (8%)
100
(1)
9982
(2 + i) =
5 e(2k+)i , = tan1
1
,
2
z = ln(2 + i) = ln( 5 e(2k+)i ) = ln 5 + (2k + )i (k = 0 , 1 , 2 )
12.
9983
12. 100
1. Find the inverse Laplace transform of the following function (10%)
F (s) =
e2s
+ 2e3s
s2 + 2s + 3
100
e2s
+ 2e3s }
2
(s + 1) + 2
1 (t2)
sin[ 2 (t 2)] H(t 2) + 2 (t 3)
= e
2
1
L {F (s)} = L {
2. Find the Laplace transform of the following function (10%)
0t<1
0
g(t) = t 1 1 t < 2
1
t2
100
g(t) = (t 1){H(t 1) H(t 2)} + H(t 2) = (t 1)H(t 1) (t 2)H(t 2)
L {g(t)} = L {(t 1)H(t 1) (t 2)H(t 2)} =
es e2s
2
s2
s
9984
3. Find the most general solution to the ODE (10%)

3
x
dy
y = xe 3
dx
100
What is the unique solution satisfying y(0) = 1 ?
dy 1
x x
y = e3
dx 3
3
I = e 3 ,
Iy =
e 3
ODE
y(x) =
x2
x x
e 3 dx =
+c
3
6
x
x2 x
e3 + c e3
6

4. Find the least-square solution of Ax = b, with A = 2 1 , b = 1 .
1 2
1
(10%)
100
1
x = (AT A)1 AT b =
3
2
2
5. Find the expression for the plane tangential to xy 3 z 2 = 4 at the point (1 , 1 , 2).
(10%)
100
= xy 3 z 2 = 4,
= y 3 z 2 i + 3xy 2 z 2 j + 2xy 3 z k
12.
9985
(1 , 1 , 2)
n = = 4 i + 12 j + 4 k
(1 , 1 , 2)
4(x 1) + 12(y 1) + 4(z 2) = 0
4x + 12y + 4z = 24
6. Solve the following problem of partial dierential equation (10%)

T
2T
+ f (x , t)
=
t
x2
100
T (0 , t) = g(t) , T (L , t) = h(t) , T (x , 0) = p(x).
T (x , t) = (x , t) +
h(t) g(t)
x + g(t) PDE
L
h (t) g (t)
2
+
x + g (t) = 2 + f (x , t)
t
L
x
2
h (t) g (t)
= 2 + f (x , t)
x g (t)
t
x
L
T (0 , t) = g(t) = (0 , t) + g(t) (0 , t) = 0
T (L , t) = h(t) = (L , t) + h(t) g(t) + g(t) (L , t) = 0
T (x , 0) = p(x) = (x , 0) +
h(0) g(0)
x + g(0)
L
h(0) g(0)
x g(0)
L
nx
},
(0 , t) = (L , t) = 0, {sin
L
(x , 0) = p(x)
(x , t) =

n=1
f (x , t)
an (t) sin
nx
L

h (t) g (t)
nx
qn (t) sin
x g (t) =
L
L
n=1
(1)
9986
(1)

n=1
an (t) sin

nx
n2 2
nx
nx
an (t)( 2 ) sin
qn (t) sin
=
+
L
L
L
L
n=1
n=1
{an (t) + (
n=1
n 2
nx
) an (t) qn (t)} sin
=0
L
L
an (t) + (
n 2
) an (t) = qn (t)
L
(2)
n 2
(2) ODE , I = e( L ) t ,
n 2
Ian (t) = e( L ) t qn (t) dt + An
( n
)2 t
L
an (t) = e
n 2
n 2
t
e( L ) qn ( ) d + An e( L )
(x , t) =
( n
)2 t
L
{e
n=1
(x , 0) = p(x)
2
An =
L
n 2
n 2
e( L ) qn ( ) d + An e( L ) t } sin
T (x , t) = (x , t) +
L
0
nx
L

h(0) g(0)
nx
An sin
x g(0) =
L
L
n=1
{p(x)
nx
h(0) g(0)
x g(0)} sin
dx
L
L
h(t) g(t)
x + g(t)
L
7. Expand f (x) = x2 for 0 < x < L. (15%)

(a) in a sine series.
(b) in a cosine series.
(c) in a Fourier series.
100
(a) Fourier sine series : f (x) L < x < L 2L ,

f (x) =

n=1
bn sin
nx
L
12.
9987
2
bn =
L
f (x) =
x2 sin
bn sin
n=1
2L2 [2 + (2 n2 2 ) cos n]
nx
dx =
L
n3 3
2L2 [2 + (2 n2 2 ) cos n]

nx
nx
=
sin
3
L
(n)
L
n=1
(b) Fourier cosine series : f (x) L < x < L 2L ,
f (x) = a0 +
an cos
n=1
1
a0 =
L
2
an =
L
f (x) = a0 +

n=1
L
0
L2
x dx =
3
2
nx
(1)n 4L2
4L2 cos n
=
dx =
L
n2 2
(n)2
x2 cos
nx
L
an cos
(1)n 4L2
nx
nx
L2
cos
=
+
2
L
3
(n)
L
n=1
(c) Fourier series : f (x) 0 < x < L L ,

f (x) = a0 +
(an cos
n=1
1
a0 =
L
2
an =
L
2
bn =
L
L
0
x2 dx =
L2
3
x2 cos
2nx
L2
dx = 2 2
L
n
x2 sin
2nx
L2
dx =
L
n
2nx
2nx
+ bn sin
)
L
L
2nx
2nx
+ bn sin
)
L
L
n=1
L2
L2
2nx L2
2nx
=
{ 2 2 cos
+
sin
}
3
L
n
L
n=1 n
f (x) = a0 +
(an cos
9988
8. Evaluate the following integrals, where C is the circle x2 + y 2 = 4 (counterclockwise)

z dz (8%)
(a)

C
1
100
z 2 (ez + e z ) dz (8%)
(b)
C
(a) Greens

z
z dz = 2i
dxdy = 2i 22 = 8 i
dxdy = 2i
z
C
D
D
1
(b) f (z) = z 2 e z , f (z) z = 0 , f (z) z = 0 Laurents
f (z) = z 2 (1 +
Resf (0) =

1
1
= ,
3!
6
2
z (e
1
z
z 2 e z dz = 2 i Resf (0) =
+ e ) dz =
1
1 1
1 1
+
+ )
+
2
z 2! z
3! z 3
i
2i
=
6
3
9. Show that cosh x sin y is a harmonic function and nd its conjugate harmonic
function. (9%)
100
(a)
2
2
(cosh
x
sin
y)
+
(cosh x sin y)
x2
y 2
= cosh x sin y cosh x sin y = 0
2 (cosh x sin y) =
cosh x sin y harmonic
12.
9989
(b) v(x , y) cosh x sin y conjugate harmonic ,
= (cosh x sin y) = cosh x cos y
x
y
v
=
(cosh x sin y) = sinh x sin y
y
x
v = sinh x cos y + f (y)

v = sinh x cos y + g(x)
v(x , y) = sinh x cos y + c
9990
13. 100
1. Find the transient current if R = 6, L = 1H, C = 0.04F ,
E = 600(cos t + 4 sin t)V ; (L, R, C, E, are measured in henrys, ohms, farads,
volts, respectively.) Initial current and charge are assumed to be zero. (10%)
100
q(t) , i(t),
L
1
d2 q
dq
+ R + q = E(t)
2
dt
dt C
d2 q
dq
+ 6 + 25q = 600(cos t + 4 sin t)
2
dt
dt
dq
q(0) = 0, (0) = 0
dt
(1) : q(t) = emt (1)
m2 + 6m + 25 = 0 m = 3 4i
qh (t) = e3t (c1 cos 4t + c2 sin 4t)

(2) :
1
600(cos t + 4 sin t)
+ 6D + 25
1
= 600 2
(cos t + 4 sin t)
1 + 6D + 25
D4
(cos t + 4 sin t)
= 600
6(D + 4)(D 4)
qp (t) =
D2
(1)
13.
9991
D4
(cos t + 4 sin t)
D 2 16
100
( sin t + 4 cos t 4 cos t 16 sin t)
=
17
= 100 sin t
= 100
(3) :
q(t) = qh (t) + qp (t) = e3t (c1 cos 4t + c2 sin 4t) + 100 sin t
q(0) = 0 = c1
c1 = 0 , c2 = 25
q (0) = 0 = 3c1 + 4c2 + 100
q(t) = 25e3t sin 4t + 100 sin t
i(t) =
dq
= 75e3t sin 4t 100e3t cos 4t + 100 cos t
dt
2. (10%) Consider a general nonhomogeneous linear ODE

y (n) + pn1 (x)y (n1) + pn2 (x)y (n2) + + p1 (x)y + p0 (x)y = r(x)
The particular solution yp (x) can be solved by the method of variation of parameters. That is
yp (x) =
yk (x)
k=1
Wk (x)
r(x) dx
W (x)
(Where the yk s are n linearly independent homogeneous solutions. W is the

Wronskian of y1 , , yn , and Wk is identical to Wi but with the kth column
replaced by a column of zeros-except for the bottom element, which is 1.) Try to
solve the following equation using the method of variation of parameters.
3
3
y + x2 y x3 y = 9x5/2
4
4
100
(1) : y = emx ODE

3
3
m(m 1)(m 2) + m = 0
4
4
9992
m=
3
1
1 ,
2
2
yh (x) = c1 x1/2 + c2 x + c3 x3/2
(2) y1 = x1/2 y2 = x y3 = x3/2 ,

yp (x) = 1 (x)y1 (x) + 2 (x)y2 (x) + 3 (x)y3 (x)

x1/2
x x3/2

1
3 1/2

1/2
1
x
W (y1 , y2 , y3 ) = 2 x
2

3 1/2
1 3/2
x
x
0
4
4

x x3/2

W (y2 , y3 ) =
3 1/2
x
1
2

1
=
4

x1/2

x3/2

1 3/2
= x
, W (y1 , y3 ) = 1
3 1/2
2
x
x1/2

2
2

=x

x1/2

x

1 1/2

= x
W (y1 , y2 ) = 1

1/2
1 2
x
2
1 (x) =
2 (x) =
3 (x)
9x5/2 W (y2 , y3 )
18
= 18x4 1 (x) = x5
W (y1 , y2 , y3 )
5
9x5/2 W (y1 , y3 )
= 36x7/2 1 (x) = 8x9/2
W (y1 , y2 , y3 )
9x5/2 W (y1 , y2 )
9
= 18x3 1 (x) = x4
=
W (y1 , y2 , y3 )
2
yp (x) = x1/2 (
18 5
9
1
x ) + x(8x9/2 ) + x3/2 ( x4 ) = x11/2
5
2
10
(3) : y(x) = yh (x) + yp (x)
3. Solve the following equation by Laplace Transform method.

y + y = f (t) , y(0) = 3

where f (t) =
0
0t<
(10%)
2 cos t t .
100
13.
9993
f (t) = 2 cos t H(t ) ,

s
L {f (t)} = L {2 cos t H(t )} = 2e L {cos(t + )}
s
= 2es L { cos t} = 2es ( 2
)
s +1
ODE LT
sY (s) y(0) + Y (s) = L {f (t)}
(s + 1)Y (s) = 3 2es (
Y (s) =
s2
s
)
+1
2s
1
s + 1 s
3
3
s
e
+
(
)e
=
s + 1 (s + 1)(s2 + 1)
s+1
s + 1 s2 + 1
y(t) = L 1 {Y (s)}
= 3et + {e(t) cos(t ) sin(t )}H(t )
4. (10%) Find the basis of solution y(x) of the following dierential equation. Show
the details of your work
xy + (2x + 1)y + (x + 1)y = 0
100
{xD 2 + (2x + 1)D + (x + 1)}y = 0 D =
(xD + x + 1)(D + 1)y = 0
z = (D + 1)y,
(xD + x + 1)z = 0 xz + (x + 1)z = 0
d
dx
9994
dz
x+1
=
dx
z
x
ln |z| = x ln |x| + c1
z = c1
ex
, z = (D + 1)y ,
x
y + y = c1
I = ex ,
Iy =
ex c1
ex
x
ex
dx = c1 ln |x| + c2
x
ODE
y(x) = c1 ex ln |x| + c2 ex
5. Find a unit vector normal to surface S given bye cos(xy) = ez 1 at the point
(1 , , 0). (10%)
100
= cos(xy) ez + 1 = 0 ,
= y sin(xy) i x sin(xy) k ez k
(1 , , 0) = k
n =

(1 , , 0)
=k
|(1 , , 0)|
6. Let F = (x y) i + (y z) j + (z x) k . Evaluate the surface integral of F

over the unit sphere dened by x2 + y 2 + z 2 = 1. (10%) 100
S : x2 + y 2 + z 2 = 1 , D,
F
n dA =
( F ) dxdydz =
3 dxdydz = 3
D
4
= 4
3
13.
9995
7. Dene the Fourier transform of f (x) to be f (w) =

f (x)eiwx dx.
2
a |x| |x| < a
(a) (5%) Calculate the Fourier transform of f (x) =
.
0
otherwise
(b) (5%) Consider the one-dimensional diusion equation :
2
u(x , t) = D 2 u(x , t) for < x <

t
x
with the initial condition u(x , 0) = f (x). Use the Fourier transform to show
that the solution of the diusion equation takes the form

K(x , t)f () d
u(x , t) =
Find K(x ), which is called
the kernel.
x2
exp( 2 ) dx = 2 2 )
( Hint : Caussian integral
2
100
(a) f (x) ,
F {f (x)} =
=
=
=

1
f(w) =
f (x)eiwx dx
2
f (x)(cos wx i sin wx) dx

2
f (x) cos wx dx
2 0

a
2
2 1 cos aw
(a x) cos wx dx =
0
w2
(b) PDE FT
dU(w , t)
= D(w 2 )U(w , t)
dt
F {u(x , t)} = U(w , t),

dU
+ w2D U = 0
dt
9996
U(w , t) = Aew
2 Dt
U(w , 0) = F {f (x)} = F (w) = A,

U(w , t) = F (w)ew
w 2 Dt
{e
} =
=
=
=
2 Dt

1
2
ew Dt eiwx dw
2

2
1
Dt(w 2 iwx+( ix )2 x
2 Dt
e
e 4Dt dw
2

1 x2
Dt(w ix )2
4Dt
2 Dt
e
e
dw
2

1 x2
e 4Dt
Dt
2
1
2
} = F 1 {F (w)} F 1 {ew Dt }
2

2
x
1
1
= f (x) e 4Dt
Dt
2
2

(x)2
1
f () d
e 4Dt
=
2
Dt

1 (x)2
e 4Dt
K(x ) =
2
Dt
u(x , t) = F 1 {F (w)ew
2 Dt
8. (a) (5%) Find and classify all local maxima, local minima and saddles for
f (x , y , z) = exp(2x2 + xz 5z 2 )
(b) (5%) Consider a forced vibration system which is described by the equations
2
d x1
2 + 2x1 x2 = A sin(wt)
dt
d2 x2
x1 + 2x2 = B sin(wt)
dt2
where A, B and w are constant. To seek a particular solution, we assume
x1 (t) = q1 sin wt and x2 (t) = q2 sin(wt). Find q1 and q2 .
100
13.
9997
(a) f y , f (x , y , z) = f (x , z) ,
= exp(2x2 + xz 5z 2 ) (4x + z) = 0
= exp(2x2 + xz 5z 2 ) (x 10z) = 0
z
x = z = 0, (0 , 0) ,
2f
2f
2f
(0 , 0) = 1
(0
,
0)
=
4
,
=
10
,
x2
z 2
xz

fxx (0 , 0) fxz (0 , 0) 4
1

=
= 41 < 0
D=
fxz (0 , 0) fzz (0 , 0) 1 10
(x , z) = (0 , 0) saddles point
(b) x1 (t) = q1 sin wt x2 (t) = q2 sin wt ODE

w 2 q1 + 2q1 q2 = A
w 2 q2 q1 + 2q2 = B
q1 =
2A B + Aw 2
A 2B + Bw 2
,
q
=
2
3 4w 2 + w 4
3 4w 2 + w 4
9. (12%) Along the circumference of the circle r = b a solution T (r , ) of Laplaces

equation is required to take on the value T0 when 0 < < and the value T0
when < < 2. Determine an expression for T valid when r > b. (Show the
details of your work)
1 2T
1 T
(r
)+ 2 2
T =
r r r
r
2
100
, T (r , ) = T (r , + 2) ,

1 , cos n , sin n
n=1
,
T (r , ) = A0 (r) +
{An (r) cos n + Bn (r) sin n}
n=1
9998
PDE
A0 (r) +
n=1

1
+ A0 (r) +
r
n=1
1
{(n2 )An (r) cos n + (n2 )Bn (r) sin n} = 0
+ 2
r n=1

1
1
n2
{A0 (r) + A0 (r)} +
[An (r) + An (r) 2 An (r)] cos n
r
r
r
n=1

2
1
n
+[Bn (r) + Bn (r) 2 Bn (r)] sin n = 0
r
r
1
A
A0 (r) = 0
(r)
+
1
n2
An (r) + An (r) 2 An (r) = 0
r
r
Bn (r) + 1 Bn (r) n Bn (r) = 0

r
r2
r 2 A0 (r) + rA0 (r) = 0
r 2 An (r) + rAn (r) n2 An (r) = 0
2
r Bn (r) + rBn (r) n2 Bn (r) = 0
(1) EulerCauchy ,
A0 = a0 + b0 ln r
An = an r n + bn r n
Bn = cn r n + dn r n
T (r , ) = a0 + b0 ln r +
{(an r n + bn r n ) cos n + (cn r n + dn r n ) sin n}
n=1
lim T (r , ) = bounded b0 = bn = dn = 0
T (b , ) = a0 +
{an bn cos n + cn bn sin n}
n=1
1
a0 =
2
1
{
T (b , )d =
2
T0 d +
0
(T0 ) d} = 0
(1)
13.
1
an b =
9999
1
T (b , ) cos n d = {
cn b
T0 cos n d +
0
(T0 ) cos n d} = 0
1 2
=
T (b , ) sin n d
0

1
=
T0 sin n d +
(T0 ) sin n d}
{
0
2T0 (1 cos n)
=
n
T (r , ) =
2T (1 cos n)

0
r n sin n
nbn
n=1
10. (Complex analysis) Prove the following identity (8%)

1
sin1 z + cos1 z = (4n + 1) , n = 0 , 1 , 2 ,
2
100
w1 = sin1 z w2 = cos1 z,
z = sin w1 = cos w2
w1 = 2n +
w2 (n = 0 , 1 , 2 ) ,
2
sin1 z + cos1 z = w1 + w2 = 2n +
w2 + w2 = 2n + (n = 0 , 1 , 2 )
2
2
99100
14. 100
1. Find the general solution of 2y
d3 y
dy d2 y
dy 2
+
2(y
+
3
+
2(
)
) = sin x.
dx3
dx dx2
dx
100
ODE
d
d2 y
dy d2 y
dy d2 y
dy 2
(2y 2 ) 2
)
) = sin x
+
2(y
+
3
+
2(
dx
dx
dx dx2
dx dx2
dx
d2 y
dy
dy dy 2
d
d2 y
= sin x
(2y 2 ) + 2y 2 + 2( )2 + 4
dx
dx
dx
dx
dx dx2
d
d2 y
dy
d dy
d
(2y 2 ) + (2y ) + 2 ( )2 = sin x
dx
dx
dx
dx
dx dx
2y
dy 2
d2 y
dy
+
2(
) = cos x + c1
+
2y
dx2
dx
dx
d
dy
d
(2y ) + (y 2 ) = cos x + c1
dx
dx
dx
2y
u = y 2,
dy
+ y 2 = sin x + c1 x + c2
dx
du
dy
= 2y ,
dx
dx
du
+ u = sin x + c1 x + c2
dx
I = ex ,
Iu(x) =
=
ex ( sin x + c1 x + c2 ) dx
1
(ex sin x + ex cos x) + c1 (xex ex ) + c2 ex + c3
2
1
u(x) = y 2 (x) = ( sin x + cos x) + c1 (x 1) + c2 + c3 ex
2
14.
99101
2. Use Frobenius method to nd the general solution of x2 y + 6xy + (6 4x2 )y = 0.

(20%)
100
x = 0 ,
y=
an xn+r
(0 < |x| < , a0 = 0)
n=0
ODE
x2
(n + r)(n + r 1)an xn+r2 + 6x
n=0
+6
n=0
an xn+r 4x2

n=0
an xn+r = 0
[(n + r)2 (n + r) + 6(n + r) + 6]an xn+r
n=0
(n + r)an xn+r1
n=0
(n + r + 2)(n + r + 3)an xn+r
n=0

n=0
4an xn+r+2 = 0
4an2 xn+r = 0
n=2
(r + 2)(r + 3)a0 xr + (r + 3)(r + 4)a1 x1+r

+ [(n + r + 2)(n + r + 3)an 4an2 ]xn+r = 0
n=2
(r + 2)(r + 3)a0 = 0
(1)
(r + 3)(r + 4)a1 = 0
(2)
(n + r + 2)(n + r + 3)an 4an2 = 0 (n = 2 , 3 , )
(3)
a0 = 0, (1) r = 2 3, (2) a1 = 0,
an =
4
an2
(n + r + 2)(n + r + 3)
(n = 2 , 3 , )
99102
4
a0
(r + 4)(r + 5)
n=2
a2 =
n=3
a3 = 0
n=4
a4 =
n=5
a5 = 0
4
42
a2 =
a0
(r + 6)(r + 7)
(r + 4)(r + 5)(r + 6)(r + 7)
a2n+1 (r) = 0,
a2n (r) =
4n
a0
(r + 4)(r + 5) (r + 2n + 3)
(1) r = 3,
4n
4n
a0 =
a0 ; a2n+1 (3) = 0
a2n (r = 3) =
1 2 3 (2n)
(2n)!
y1 (x) =

n+r
an x
n=0
r=3
a2n x2n3 +
n=0
a2n+1 x2n2
n=0
2n
(2x)

4n 2n
x = a0 x3
n=0 (2n)!
n=0 (2n)!
3
= a0 x cosh 2x
= a0 x3
(2) r = 2,
a2n (r = 2) =
4n
4n
a0 =
a0 ; a2n+1 (2) = 0
2 3 4 (2n + 1)
(2n + 1)!
y2 (x) =

n=0

an xn+r
r=2

n=0
a2n x2n2 +

n=0
a2n+1 x2n1
(2x)2n+1

4n
a0
x2n = x3
= a0 x2
2
n=0 (2n + 1)!
n=0 (2n + 1)!
a0 3
x sinh 2x
=
2
y(x) = c1 y1 (x) + c2 y2 (x) = c1 x3 cosh 2x + c2 x3 sinh 2x
14.
99103
3. (15%) Find the eigenvalues, eigenfunctions and verify the orthogonality of the
solution by direct calculation.
y + y = 0 , y() = y() , y () = y ()
100
(a) < 0 ; = p2 (0 < p < ) , ODE

y (x) p2 y(x) = 0
ODE
y(x) = c1 epx + c2 epx
(1)
y (x) = p c1 epx p c2 epx
(2)
y() = y() (1) ,

c1 ep + c2 ep = c1 ep + c2 ep
c1 (ep ep ) + c2 (ep ep ) = 0
(ep ep ) = 0,
c 1 c2 = 0
(3)
y () = y () (2) ,

p c1 ep p c2 ep = p c1 ep p c2 ep
c1 (ep ep ) c2 (ep ep ) = 0
(ep ep ) = 0,
c 1 + c2 = 0
(4)
(3) (4) c1 = c2 = 0, c1 = c2 = 0 (1) , y(x) = 0 ,

ODE
99104
(b) = 0 , y (x) = 0 , ODE

y(x) = c1 x + c2 y (x) = c1
(5)
y() = y() (5) ,

c1 () + c2 = c1 + c2
c1 = 0 , y () = y () (5) , c1 = c1 , ,
c2 , c2 = 0 , = 0 ODE ,
y(x) = c2 (c2 = 0)
(c) > 0 ; = p2 (0 < p < ) , ODE
y (x) + p2 y(x) = 0
ODE
y(x) = c1 cos px + c2 sin px
(6)
y (x) = p c1 sin px + p c2 cos px
(7)
y() = y() (6) ,

c1 cos(p) + c2 sin(p) = c1 cos p + c2 sin p
c2 sin p = 0
(8)
y () = y () (7) ,

p c1 sin(p) + p c2 cos(p) = p c1 sin p + p c2 cos p
c1 sin p = 0
(8) (9) c1 c2 0 , sin p = 0,

p = n (n = 1 , 2 , 3 , 4 )
ODE = p2 = n2 (n = 1 , 2 , 3 , 4 )
y(x) = c1 cos nx + c2 sin nx (c1 c2 0)
(9)
14.
99105
4. A surface is dened by z = x2 + y 2, calculate the surface are dened on the domain

100
0 x2 + y 2 4. (15%)
S = x2 + y 2 z = 0,
2x i + 2y j k
=
n =
||
4x2 + 4y 2 + 1

dx dy
dA =
4x2 + 4y 2 + 1 dx dy
=

|k n|
A =

4x2 + 4y 2 + 1 dx dy
dA =
S
Sxy
Sxy S xy
=2
r=2
=
4r 2 + 1 r dr d
=0
r=0
1
(1 + 17 17 )
=
6
5. f (x) = x, x , f (x) = f (x + 2), solve

(1) Fourier series, (10%)
100
(2) complex Fourier series. (10%)
(1) f (x) ,
f (x) =

n=1
2
bn =
bn sin nx
x sin nx dx =
2 cos n
n
99106
(2)
f (x) =
n=
1
cn =
2
cn ei nx
1
cos n
f (x)e
dx =
x sin nx dx =
0
n

1
1
c0 =
f (x) dx =
x dx = 0
2
2
i nx
(n = 0)
6. Solve nonhomogeneous boundary condition of hear equation

2 u(x , t)
u(x , t)
= c2
t
x2
u(0 , t) = A1 , u(L , t) = A2 , u(x , 0) = f (x), where 0 < x < L, t > 0.
100
u(x , t) = (x , t) +
A2 A1
x + A1 , PDE
L
2
= c2 2
t
x
(1)
u(0 , t) = (0 , t) + A1 = A1 (0 , t) = 0
u(L , t) = (L , t) + A2 A1 + A1 = A2 (L , t) = 0
u(x , 0) = (x , 0) +
A2 A1
A2 A1
x + A1 = f (x) (x , 0) = f (x)
x A1
L
L
(x , t) =

n=1
an (t) sin
nx
L
(1)

n=1
an (t) sin

nx
n2 2
nx
= c2
an (t)( 2 ) sin
L
L
L
n=1
{an (t) + (
n=1
nc 2
nx
) an (t)} sin
=0
L
L
14.
99107
an (t) + (
nc 2
) an (t) = 0
L
an (t) = Bn e(
(x , t) =

n=1
(0 , t) = f (x)
2
Bn =
L
L
0
Bn e(
nc 2
) t
L
nc 2
) t
L
sin
nx
L

A2 A1
nx
x A1 =
Bn sin
L
L
n=1
{f (x)
A2 A1
nx
x A1 } sin
dx
L
L
PDE
u(x , t) = (x , t) +
A2 A1
x + A1
L
99108
15. 100 A
1. (Total 15%) Let Mnn (C) be the vector space consisting of all n n matrices with
complex entries. Two matrices A, B Mnn (C) are said to be unitarily equivalent
if there exists a unitary matrix P Mnn (C) such that A = P BP , where P is
the conjugate transpose of P
(a) (10%) Let A, B Mnn (C) be unitarily equivalent. Show that
tr(A A) = tr(B B).

1
2
1 + 2i 1
(b) (5%) Determine whether the matrices A =
and B =
2+i 3
4i
2
are unitarily equivalent (you need to justify your answer ) ?
100 A
(a)
A A = (P BP ) (P BP ) = P B P P BP = P B BP
tr(A A) = tr(P B BP ) = tr(P P B B) = tr(B B)

(b) tr(A A) = 12 + 22 + |2 + i|2 + 32 = 19 , tr(B B) = |1 + 2i|2 + 12 + |4i|2 + 22 = 26 ,
tr(A A) = tr(B B), A B unitarily equivalent
2. (Total 10%) Let P2 (R) be a vector space that consists of all polynomials with real
coecients and with degree less or equal to 2. Let T be a linear operator on P2 (R)
dened by
T f (x) = f (2x 1) 2xf (x)
for all f (x) P2 (R)
(a) (6%) Suppose that = {1 + x2 , x + x2 , 1 + x + x2 } is an ordered basis for
P2 (R). Find the matrix representation of T in , i.e., [T ]
(b) (4%) Let A = [T ] and let U be a linear operator on R3 dened by U(x) = Ax
for all x R3 . Find a basis for the range space of U .
100 A
15. A
99109
(a) P2 (R) = {1 , x , x2 } ,
T (1) = 1 2x(0) = 1
T (x) = 2x 1 2x(1) = 1
T (x2 ) = (2x 1)2 2x(2x) = 1 4x
1 1 1
[T ] = 0 0 4
0
[T ] = [I] [T ] [I]
1 0 1
1 1 1
1 0 1
= 0 1 1 0 0 4 0 1 1
1 1 1
1 1 1
4
4
4
= 2 0 1
2 4 3
(b) R (U) = CS (A), rank (A) = 2, U
2 , 0
2
4
3. (a) (6%) Solve y +
1
y = 3x2
x
(b) (14%) Given the dierential : x(t) + a x(t)

+ b x(t) = u(t) where u(t) is a
unit-step function, the response x(t) is expressed as :
x(t) = 0.01 {1 c e5
2t
sin(5 2t + )}
All initial conditions are zero. Please calculate the constants a, b, c, and the
angle (the unit is radian).
100 A
99110

1
dx
x
= eln |x| = x,
3
I(x)y = x 3x2 dx = x4 + c
4
(a) ODE I(x) = e
3
c
y = x3 +
4
x
(b)
x(t) = 0.01 {1 c e5
2t
sin(5 2t + )}
ODE xp (t) = 0.01,
xh (t) = 0.01c e5
2t
sin(5 2t + )}
(1)
x(t) = emt
x(t) + a x(t)
+ b x(t) = 0
m2 + am + b = 0
(1) , m = 5 2 5 2 i,
a = {(5 2 + 5 2 i) + (5 2 5 2 i)} = 10 2
b = (5 2 + 5 2 i)(5 2 5 2 i) = 100
ODE
x(t) = 0.01 + e5
2t
(c1 cos 5 2 t + c2 sin 5 2 t)
x(0) = 0 = 0.01 + c1
x(0)
= 0 = 5 2 c1 + 5 2 c2
c1 = c2 = 0.01 ,
x(t) = 0.01 0.01e5 2 t (cos 5 2 t + sin 5 2 t)
1
1
= 0.01{1 2 e5 2 t ( cos 5 2 t + sin 5 2 t)}
2
2
52 t
(sin cos 5 2 t + cos sin 5 2 t)}

= 0.001{1 2 e
4
4
52 t
sin(5 2 t + )}
= 0.001{1 2 e
4
c=
2=
4
(2)
15. A
99111
4. If a periodic function whose Laplace transform is shown as :

F (s) =
s
(s2 + 1)(1 es )
(a) (10%) Using power series to expand
1
and nd the corresponding pe1 es
riodic function f (t).

(b) (5%) Please plot this function f (t) in tdomain.
(a) L 1 {
100 A
s
} = cos t ,
(s2 + 1)
f (t) = L 1 {F (s)} = L 1 {
1
s
}
(s2 + 1) (1 es )
s
(1 + es + e2s + )}
+1
= cos t + cos(t )H(t ) + cos(t 2)H(t 2) +
= L 1 {
s2
(b)
2
2
1
2
10
12
99112
5. For a function : f (x) = | sin x|, where < x < . If we wish to use a function
g(x), which is in a nite-dimensional vector space V spanned by trigonometric
functions; sin nx and cos nx for n = 0 to 5, to approximate f (x).
(a) (6%) Does the set
{sin x, sin 2x, sin 3x, sin 4x, sin 5x, cos 0x, cos x, cos 2x, cos 3x, cos 4x, cos 5x}
form an orthogonal basis in V for < x < ?
(b) (9%) Please nd g(x) in V that is closest to f (x), i.e., g(x) is with minimum
square error from f (x) in V .
You may need the following formulas :
1
1
sin x cos y = [sin(x + y) + sin(x y)] , cos x sin y = [sin(x + y) sin(x y)]
2
2
1
1
cos x cos y = [cos(x + y) + cos(x y)] , sin x sin y = [cos(x y) cos(x + y)]
2
2
100 A
(a)
< sin nx , cos mx >=
sin(nx) cos(mx) dx = 0
sin(nx) sin(mx) dx
< sin nx , sin mx > =

1

{n
cos(nx)
sin(mx)
+
m
sin(nx)
cos(mx)}

2
2
n m
= 0 (n = m)
cos(nx) cos(mx) dx
< cos nx , cos mx > =

1

{n
sin(nx)
cos(mx)
m
cos(nx)
sin(mx)}

n2 m2
= 0 (n = m)
< sin nx , sin nx >=
sin2 (nx) dx =
15. A
99113
< cos nx , cos nx >=
cos2 (nx) dx =
V
(b) f (x) = | sin x| ,
f (x) = a0 +
2
a0 =
an =
f (x) =
g(x) =

n=1
an cos(2nx)
sin xdx =
sin x cos(2nx)dx =
4
(1 4n2 )
4
2
+
cos(2nx)
n=1 (1 4n2 )
4
4
2
cos 2x
cos 4x
3
15
6. Evaluate the integrals (counterclockwise)

100 A

ez
dz , C : |z| = 4.5
(a) (10%)
C cos z
c+i
2kz
1
F (z) dz , where F (z) = 2
and k is a constant.
(b) (15%)
2i ci
(z + k 2 )2
ez
2n + 1
, cos z = 0, z =
(n = 0 , 1 , 2 ) , f (z)
cos z
2
poles , z1 = z2 = C ,
2
2
ez
Resf ( ) =
= e 2
2
sin z z= 2
(a) f (z) =
ez
Resf ( ) =
= e 2
2
sin z z= 2
99114
f (z)dz = 2i {Resf ( ) + Resf ( )} = 2i{e 2 + e 2 }

2
2
C
(b) c > 0 ,
1
2i
c+i
F (z)dz =
ci
=
=
=
c+i

2kz
1

zt
e dz
2
2
2
2i ci (z + k )
t=0
c+i

2ks
1

est ds
2
2
2
2i ci (s + k )
t=0

2ks

1
}
L { 2
2
2
(s + k ) t=0

2 sin(kt) cos(kt) = 0
t=0

100 Math 1

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

100 Math 1

Cargado por

Copyright:

Formatos disponibles

100

y(x) = c1 x + c2 x ln x + c3 x(ln x)2

2. (15%) Knowing L {J0 (t)} =

L {ty(t) + y (t) + ty(t)} = 0

lim sY (s) = lim

A = 1 cos i sin = 1 (A I)X = 0

= cos + i sin (A I)X = 0

y(x , 0) = f (x) for 0 x L,

(1) y(0 , t) = y(L , t) = 0,

an (t) sin

(4) y(x, , t) = (x + mt) PDE ,

y(x , 0) = f (x) f (x) = (x) + (x)

g() d = (x) (a) (x) + (a)

g() d = 2(x) (a) + (a)

g()d + (a) (a)}

g() d = 2(x) + (a) (a)

g()d (a) + (a)}

(5) (6) (1) , DAlembert

f (x) = H(x L) H(x L) g(x) = 0,

; 0 < x < L/6

radius 2 centered at the origin. (

dA over the circular disk S.

(3) (x0 , y0 ) C , C , (x0 , y0 ) = (0 , 0),

Resf (0) = lim

2. (20%) If f (x) is a periodic function with period of T , prove that

f (x) sin nw0 x dx

< f (x) , f (x) > =

an ](t) cos

5. (20%) Consider the following two initial values problems :

= 0, (t) : Dirac Delta function.

+ c{sY (s) y(0)} + kY (s) = 1

+ c{sY (s) y(0)} + kY (s) = R(s)

Let P = [ m | n | t ] and = 0 1 0 be the 3 3 matrices containing the

(a) A2 = A , det(A2 ) = {det(A)}2 = det(A), det(A) = 0 det(A) = 1

2. (a) (8%) Given a 1st order system ODE :

(a) det(A I) = 0, = 2 2 , = 2 (A I)V = 0

L {y(x)} = Y (s) L {f (x)} = F (s),

3. The height of a hill(in meter) is given by

h(1 , 1) = 220 i + 220 j , (1 , 1) slope

|h(1 , 1)| = 220 2

4. (10%) Evaluate the integral

5. The one dimensional wave equation

an (t) sin

y = (3x + 1)m , ODE

y(x) = c1 (3x + 1)2 + c2 (3x + 1) 2

2. Laplace Transformation (15%)

3. yex dx + (2y + ex )dy = 0. (10%)

(1) : y = emx ODE

yh (x) = c1 e2x + c2 xe2x + c3 x2 e2x

(3) : y(x) = yh (x) + yp (x)

r (t) = x(t) i + y(t) j (6%)

x(t) = c1 e4mt , y(t) = c2 e2mt

r (t) = x(t) i + y(t) j = 4e2t i + 2et j

(2) u(r , 0) = 0 , u(r , ) = 0 ,

an (r) sin(4n) +

{r 2 an (r) + r an (r) 16n2 an (r)} sin(4n) = 0

r 2 an (r) + r an (r) 16n2 an (r) = 0

i2 (0) = 0 (variation of parameters)

(2) L1 = L2 = 1, R1 = 8 R2 = 3 E = 100 sin t (1)

det(A I) = 0, = 2 12, = 2 (A I)X = 0

ODE ODE (fundamental matrix)

Ip = Q Y + QY , (1) ( ODE)

4e2t sin t 2e2t cos t

8. F = (16y + sin x2 ) i + (4ey + 3x2 ) j C