0% found this document useful (0 votes)
6 views22 pages

Transform and Integral Equation

The document discusses the properties of Fredholm integral equations, including the existence and uniqueness of solutions for both homogeneous and inhomogeneous cases. It introduces concepts such as eigenvalues and eigenfunctions, as well as singular integral equations, and presents various examples and theorems related to these topics. Additionally, it covers convolution theorems and provides methods for evaluating specific integral equations.

Uploaded by

djagannath033
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
6 views22 pages

Transform and Integral Equation

The document discusses the properties of Fredholm integral equations, including the existence and uniqueness of solutions for both homogeneous and inhomogeneous cases. It introduces concepts such as eigenvalues and eigenfunctions, as well as singular integral equations, and presents various examples and theorems related to these topics. Additionally, it covers convolution theorems and provides methods for evaluating specific integral equations.

Uploaded by

djagannath033
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 22

e Fredholm 4Hernalive lheo Yen

Ether the integral equation, y()= ftot / k(,t) y)dt -. (0)


with fixed Apossesses one and only one soluion y() for
arbitrary ds- funetions f() and k(a,t), in porticular the
Solution y=0 for f0 i or the homogeneous equation
y) = /Ko,t) yt) dt (i)
posesses a a fiite number r off (inearly odependent solutions
Jn the first case,the transþo sedd
înhomogeneou s equatio,
Z()= fe) + 2/k(t) z) dt
also possesses a unique solution. Jn the seeond ea se, the
transposed homoge neous equa ion
z)= A k(t,3) Z) dt
also has r bineorly independent solu tions Z.i , is 1, 2, , r;
the in homczeneous integral equation ) has a solution
iif the given funclion foy satisfies the r conditions
|f Zoi ()dx =0, iz,2,.",r
In this case the solution of GJ is de ter mined only upto
an additive ineor combination
E'gen fune lion
Consider the homogzneous Fredholm integral equation of
2nd Kind
k(,t) y () dt
- - . .

Y() =
let Ao be an eigen value of and oy be the

Corresponding non-trivial of ), then the funetion


) is called eigen funcion or charoeterlstie funetion
of G) corresponding to A= do
Singular Jteyol Equation An Jhtegral equation is said
+o be singular if one oY both imits of integral are
in finite.
() the kernel become ufinite at one
oY more points.

Exomple :
X) =2t + 6 sin(t-s) x(s) ds
X(S) ds
t-s

t x(s)ds
-s)
o(et) F()= write also equivalenty,
we
’ t as
oYderexponential funetion
of a îsF() thot,
|eF<m soy we
F(meot
t>t, al for that such numbert finite
aand number
o ConsBant
an mER exist there 1f
rder onential exp unetion
of
) t e tSint dt
Q. Evaluate : )J(eae) dt

)Lte sin st }
-at
Ans

-at -bt

let, f) =

Now, Leat
St b - fs) (sor)
f( dx

f) dt = X+a tb

Je()ude -(n (z+a)- In(at)


-st

S
-at

t
St
-at -bt
e(ee)stdt = ln (1) + lu(a(stb
t Stay
-st -at - a
stb
t

laking timit as then we get


-at -bt
e
t e SiMt dt

hot, L{ sint , =
We know that,

No w,
Ltsiat =-
- () 2S
(s+1)
by defination of Lap laee troms form
L{t sint}= J e(t sin t) dt
25
e t sin t) dt = (s*+I)*
them We get

[tet sint
Sint t = 2x3
(3*+)*
3
50

r(4+)
=+ 2 4 3 r<)
S

-t 2 r ) , ,3 r(4)

+ 3

() ds
d
St3

s)

(s+ 3)4
Lfesst}
3
(sH1)*+ 3°) (by using shifting proparby)
6(st)
(s+ 2s+ 1o)

() show that Ft) =l (s a exponential oroder ?.


Wshow that ue fune tion F() =et is not ot exponential

Ans!
Definition Recap! 4 funeton F) is said to be af
exponentiol order a f there eist
posifve constants , 0, lo
Sueh tthat,
F)| <meot
for all t>to
e

To check f F(t) =* is of exponenial order 2


we om pute:
lim e F )
-3t
t
e
Lim ce)
(using L'Hô pitals rale )
2t
3e3t

t (et) using L'H6Pitals rule )


SE (3e)
2

Sinee, me Fl =0 FC)-*
F)=t is of ex ponen tial orer
tlere, F()= e

we hae, F)

tlt-)
for all volues of
tenee, Whatever be e volue af we canMot determi ne a real
Constont m Sueh that

lhaye fore, the iven fmetion is not of


exponential order as t o
Q. Show thait the ho oeneous,integral aquh
do es not honWe real eigen
eigen funcion. values

Ans!
Tle given JE is Yo) = A (ta- xE) Y4) dE
oY, Y) =

where,
-(1)

Now, C,

oY, C

G(I-2h ) + 4 (2) - 0 -)
Now, = JE Y) dt

e, tJt t

Gc, (4)- (1+ 2A) =0 (ö)


for non-Zero
no solutio,
w
(1-22) 3

2
-(142)
-J+ 44
25

- 50

Henee, eigen values re not real values,


for, A= isJ6 from G) we get

br, = - l5l6 from ) We got


y() =- i5/6 C, JX +

eleorly ta eigen funetion ane


a not real.
Solutio of tlomogene ous fune iou IE of
Kind wth se barable oY degenerated Kernal
Consider homogencous fune tion JE of 2nd kind with
y() = Ak() Y)
y)dt
K(x,t) isIS separable, We take

K(x,t)= f () ,(4)
from G) in Gi) we get

Y(%) = f,) () y0 dt (G)


Let,
.(t) y() dt =c; , where i=1,2,..,n

There fore, (v)

Where, C: (i=l2-., n) are to be determinate in order


to finite solulion ofi)
multiply both sides of (v) sucee ssively by ,(), ,(0,-", ,)
om d tnteqyating over the inteyval (a, b) we get
a

where
J=,2, --yh
from Cv) isJ we get
J a.(t) Yt)JE
[using w1
oY,

OY,

Similarly, -XG, apt (J-)0)ca


Ong n - (ui)

+ (I-nn) Cn 0
This is the omogencous sysBem of Lin eor equati ons.
we are to deter miyate Ci, e2, C
Determinate D(A) = |(1- 0,)
(1- an).
1f DO) #0 then the system of equ au)(1-anm)
hos only
trivial so lutio.

tlenee, equ G) becomes only trivial solu" Y() =0


toweverr" Jf D(A) =0 at last one ot the C;'s Can be
asigned arbitrarily and the remaining 's cay ke
determined aceorainaly.
here fore when D()=0, then infinite mamy solution
of the JE GJ exist.
The values of tor which D() =0 are called eigen
salues and any non-trivial solution ot G) is
anY
ealled eorres ponling eigen funeion.
Convoluation Theorem :
If L{f}= f) ond Lae)} = 2(9 then,

Where, fy*q) is called the convolution of fit) and gt


and s denote by the iategral rt

ft)* gt) J f() J(t-)du=


and this iategyol îs called Convoluation integral and also
denoted by f*)(.
t 20 J
a-aos
t< ssintinat
2a
Cosat+
4a2
stnatSin2a6 sinat 2a t 2at+ cos CoSat Cosat
(t Aosat(ta
2 2
2at' Sin sinat Cos20t)
sinat 2 du
t 20u Sin
t
at-o) COS au.- Sin rt
cosalt-u)
y =Sinau .
t
Now,
t
ale) sinat= L"i)3 =
fe) Cosat= fO L?
ta (5=) f)=
S
whre,
S Aws;
S
sin}sl b -atsin a
2la-b)Sin(4-6utbt)
utub (asin +
L(t-u)
u Cosatt.
Cos
g6)=c3c:)}=
btcos osat =f3(91= fo=
S
(3t-)+ sin
18
et (os3t)+ - (sin
3t L
{o-)| -et(sinst
-cos7t)-
3u)7 -e( os -3usin
t -3t
3
3u
singu 3
-3(t-)
rt
St3
sinst f= '{fio-
3
f:
6
-(-t)
dtyt) e-x :y) -y()
get we-from
=o/ y'u) [since,
dty4) - c y'=
4)
in
x=l puting
Yo)] [ate, -0 =c-"yedt
y'
yl)
dt y(o-) y'=()
=0 Ey() y'+(o) y'(-)
[yej'
0Y)dE=
x to from
o side, both integral
=y)+
0oY) Giivvenen O
So!":
=0 () YY(o)= with =x 2Y +y"
=0 Y() omd =0Y(o) with y=o 0y+
Y(9=0 amd 0Y(o=) with =0 AY +y"
=0y'() amd =1Y(o) with
erify .omd FJE intoaproblem Vaue boundery Convertinatingg Q.
oY, Y() = J+

t o<t<X
where, K(x,t) =

Vevify:
Y()= J+

Wiff. wat t: both side, we get


y'() = o+ Jity}dt t xy()()- oy(")(
9)

y'a) = 0 + o st +0- X Y()

y'() =
again, t . bo th Side w.r.t. X

or, y') = f'o dt +o- Y)


y'() + Y() = O

from Cwe get, (put x =0)


Y(o) = J +
oY, Y(o) =

puting x=1 in ) we 9et -

y'() =
y'(4) = O

Henee, (1) = 0 ith y(o)=1 & y'u) ==


y'() +Y)
0
implies, y(x) = Jt J't y4)dt + x y(t) Jt is verify.
() Given y + Ay =0 wth (0) = Y() = 0

integral both side, from oto x


y"() dt + Y(t) dt =0

Y(t) Jt=0

y'(%)
again interal both side fron
dt

Y() - Y(o) = -

Y() = xe - )(x-t) Et) t [ sinee,, Y(o) =o 7


putting =! we gee
y t

e = A /(-ttt) t [ since, y(4) =o]


putin a the value ot e e in 0, we get
Y() = J (x-t) y()dt
A(u- t) y dt + (4-t) y(H JEt -
oY, Y() = A t +
(-x) y()dE (l-t) Y(U dt

Y(x) = k (,t) Y) dE

Where, K (x,t) = t(l-x) o<t<X

X < t <l
Given, y'+ xy =0 wilh Y(0) =0- Y()a
inteçra both side Yom o to x

y"() JH + { y() Jt =0

Y'()-y(o) + t yt) t o

Y(*)
JtY) Jt [ot, e yt)]
agen, integral both side fron
fron o to x

t t y )dt

or,
Y()- Y(o)
Y(0) = -J(-t) Y) JE
6

Y() = xe- | (x-t) Y(t) dt -O Sine, Y(0) =o


puting x=1 in O, we 9et

(1- t) Y() Jt
or,

Y(*)= X| (4-t) y(E) t - J(a-t) Yet) dt


or, y(G)= x(1-t) YUdt +
or, y() =
Jx(4-t) y(t) Jt
Y) = J ko) Y) JE
J t(1- 2 ) : o t<x
where, K(,) = 1x(r
(1- t) ;
<t<!
te1
(,t)= K
o<t< X
(a-x) tf were,
K(3)
dtY) J1-)+ Y() or,
+ -(a) Y or,
Y)
= ors
Y()=
+
get 2 we from
6
dtYct) (7-) -J
get we in
) inX=7 putt
Y(x)=
cx=*+ Y(o) Yo)-
y')dt J
xtofromo side both integral again
|yt)
dt
2
= dt
Y) y(+
o) Y()-
ay4)
=dt Jyt)
+dt
from side,
to o
bo-h integral
Y(7)
0wihY(o)
= = X y y"+ Given,
iv)
show hort ifa fune tion f(3) define d on (-o,a)
om is Fourier tronsfonm F(S) ne boho real, hen
fo) is?s eVen. Also sho wthat if foo is Teal its
Fourier transform F(s) is purely inagjnary, hen
f) is odd
Ans,
By the definaton of Fourier trans form,
F(S) =

part J
f(oy is real ’ fo) = fy
F(S) s Yeal ’ F() = F(S)
(Sinee, Teal funetion is Camplex onju gate of itself)
Now,
F(9) = J foy e's*
-isx
fye

F(S) =
- is(-)

Or, F(S) = J fye'(su alu


J feyeisu u

f9 =f-)
foo is eve
Port 2!

fo9 is real fo) =f )


F(S) is purely imaginary ’ F() = - F(s)

NoW,
F(S)
J foy esx
-isx
foy e

Cet, =-U then,


- -is (-)
F(S) = J. fcu) e du)
(Su
-F(S) = f)e
or,

fx) = -f)
fo«) is odd

You might also like