Existence of solution legendre equation

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
1 reply · 2K views
saxen
Messages
44
Reaction score
0
Hi all, I have my exam in differential equations in one week so I will probably post a lot of question. I hope you won't get tired of me!


Homework Statement


This is Legendres differential equation of order n. Determine an interval [0 t_0] such that the basic existence theorem guarantees a solution.

(1-t^2) [itex]\frac{d^2y}{dt^2}[/itex] - 2t[itex]\frac{dy}{dt}[/itex]+n(n-1)y = 0
y(0)=0
y'(0)=0


Homework Equations



Picard-Lindelöfs theorem
M = max |f(x)|

T_0 = min {T,δ/M}

The Attempt at a Solution



I thought, Picards theorem is for first order equation so thought I should first rewrite the system as x'(t)=A(t)x(t)

Write

[itex]x_{1}[/itex] = y --- >[itex]dx_{1}/dt[/itex] = dy/dt

[itex]x_{2}[/itex] = dy/dt --- > [itex]dx_{2}/dt[/itex] = d^2y/dt^2

dx_1/dt = [itex]\frac{1}{1+t^2}[/itex]

dx_2/dt = [itex]\frac{-n(n+1)}{1+t^2}[/itex] + [itex]\frac{2t)}{1+t^2}[/itex]


Picards theorem require the function to be lipschitz, which it seems to be.

Picards theorem also states that you should pick the interval according to

M = max |f(x)|

T_0 = min {T,δ/M}

Here is where I am stuck, assuming I correct so far.

M = max |f(x)|

T_0 = min {T,δ/M}

I think M is equal to n(n+1) but I have NO idea on how to compute T_0.


As always, all help is greatly appreciated!
 
Physics news on Phys.org
saxen said:
dx_1/dt =x_2

dx_2/dt = [itex]\frac{-n(n+1)}{1+t^2}[/itex]x_1 + [itex]\frac{2t)}{1+t^2}[/itex] x_2

Some corrections to my current solution