Scribd
Upload
57 views

Engineering Analysis Homework 8

This document contains solutions to homework problems on engineering analysis. Problem 1 involves solving a second order ordinary differential equation with boundary conditions to find the eigenvalues and eigenfunctions. The eigenfunctions are shown to be orthogonal. An expression is derived for the coefficients an in terms of the eigenfunctions. Problem 2 involves another second order ODE with different boundary conditions. Part a solves for the eigenvalues and eigenfunctions. Part b proves that the eigenvalues are real, there are an infinite number of eigenvalues, and the eigenvalues are not repeated. It is also shown that as n approaches infinity, the eigenvalues λn approach infinity. Part c gives an approximate expression for the eigenvalues λn as n approaches infinity.

Uploaded by

Amrita
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
57 views

Engineering Analysis Homework 8

This document contains solutions to homework problems on engineering analysis. Problem 1 involves solving a second order ordinary differential equation with boundary conditions to find the eigenvalues and eigenfunctions. The eigenfunctions are shown to be orthogonal. An expression is derived for the coefficients an in terms of the eigenfunctions. Problem 2 involves another second order ODE with different boundary conditions. Part a solves for the eigenvalues and eigenfunctions. Part b proves that the eigenvalues are real, there are an infinite number of eigenvalues, and the eigenvalues are not repeated. It is also shown that as n approaches infinity, the eigenvalues λn approach infinity. Part c gives an approximate expression for the eigenvalues λn as n approaches infinity.

Uploaded by

Amrita
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

A homework on Engineering Analysis 8

By AMRIT ARYAL
Oct,2020

Problem 1: SL ODE
Given, Eigen value function is,

dy 2
+ λ2 y(x) = 0 (1)
d2 x
for 0<x<1

with boundary conditions y 0 (0) = 0 andy(1) = 0.

1a: The eigenvalues and eigenfunctions.


Eq(1) is second order ODE.

The solution is
y(x) = c1 sin λx + c2 cos λx

Applying the boudary condition

y 0 = λc1 cos λx − λc2 sin λx

@x = 0, y 0 = 0 0 = c1
And,

y = c2 cos λx

@x = 1,y(1) = 0

2n−1
λ= 2 π for n=1,2,3...∞. are the eigen vectors.

Corresponding eigen functions are cos 2n−1


2 πx for n=1,2,3........∞.

1
1b: The eigenfunctions are orthogonal
d dy
Compare eq(1) with standard function. dx (p(x) dx − q(x)y(x) + λr(x)y(x) = 0
p(x) = 1,q(x) = 0,r(x)=1

Orthogonal
Rb
if a r(x) cos π2 x cos 3π
2 xdx = 0
or,
R1 1
0 2
[cos( π2 + 3π
2 )x + cos( π2 − 3π
2 )x]dx
R1
or, 0 12 [cos(2π)x + cos(π)x]dx

or, 21 [2π sin(2π)x|10 + π sin πx|10 ] =0.

∴ all eigen functions are orthogonal.

1c. the coefficients an


R1
an = 0
r(x)f (x)Φn (x)dx
R1
an = 0
f (x)Φn (x)dx
R 1 P∞
an = 0 n=0 Φn (x)Φn (x)dx
R 1 P∞
an = 0 n=0 cos 2n−1 2n−1
2 πx cos 2 πxdx
is the expression for an .

2
Boundary condition is y(0) + y 0 (0) = 0 and y’(1)=0

2a.1
Eq(1) is second order ODE.

The solution is
y(x) = c1 sin λx + c2 cos λx

y 0 = λc1 cos λx − λc2 sin λx

Applying the boudary condition

@x = 0 y(0) + y 0 (0) = c2 + λc1

@x = 1 y 0 (1) = 0

2
or,λc1 cos λ − λc2 sin λ = 0

or,c1 cos λ = c2 sin λ

orc2 = c1 cot λ

or,−λc1 = c1 cot λ

λ = − cot λ.
Eigen values are the solution of λ = − cot λ and correspondings eigen function
are − cot λx.

2b.1 eigen values are real


Let λ = u + iv

Corresponding eigen functions are


Φ = U + iV

Using lagrange identity.


L(y(x) = λr(x)y(x)

or,L(Φ) = λrΦ
and
Rb Rb
a
ûL(v)dx = a vLûdx
we get,
put u=Φ̂ and v=Φ.
Rb Rb
a
Φ̂L(Φ)dx = a ΦLΦ̂dx
Rb Rb
a
Φ̂λrΦdx = a
Φλ̂rΦ̂dx

ΦΦ̂ = |Φ|2 . complex conjugate of two number is real number.


rewrite,
Rb Rb
a
λr|Φ|2 dx − a λ̂r|Φ|2 dx = 0
Rb
a
[(λ − λ̂)r dx − r|Φ|2 ]dx = 0

here,|Φ|2 ≥ 0, r(x) > 1

∴ (λ − λ̂) = 0

λ = λ̂
All eigen value are real.

3
2b.2 There are infinite eigenvalue
Solution of above equation can be write as
P∞
f (x) = n=1 Bn Φn (x)
which is square integrable.
Rb
i.e a f (x)2 r(x)dx = f inite
And,
Rb
This solution is only valid when error between two solution is zero. a |f (x) −
PN 2
n=1 Bn Φn (x)| r(x)dx → 0.if N→ ∞
∴ It has infinite eigen values.

2b.3 eigen value are not repeated


put u=Φ1 and v=Φ2 .
Rb Rb
a
λ1 rΦ1 Φ2 dx − a
λ2 rΦ1 Φ2 dx = 0
Rb
a
(λ1 − λ2 )rΦ1 Φ2 dx = 0
If (λ1 − λ2 ) = 0, trivial solutions. since (λ1 − λ2 ) 6= 0
Rb
a
rΦ1 Φ2 dx = 0 eigen function are orthogonal. Eigen function are orthogo-
nal only when λ1 6= λ2 )

2b.4 n→ ∞ λn → ∞
λn = − cot λn has distinct solution every value of n , as n → ∞ λn has infinite
solution.

2c:Expression for λn
λn = − cot λn have distinct eigen functions.
As, n → ∞, λn = − cot λn→∞ is the approximate expression for λn .

You might also like