Series Solution of an ODE: Finding a Non-Recursive Formula

Click For Summary
SUMMARY

The forum discussion focuses on solving the ordinary differential equation (ODE) y' = x²y using a power series approach. The user derives a recursion formula a_{n+1} = a_{n-2}/(n+1) for n ≥ 2 but struggles to eliminate recursion to find a non-recursive formula. Key conclusions indicate that coefficients a_1 and a_3 must equal zero, leading to the realization that a_2 serves as the arbitrary constant in the solution. The final insight clarifies that non-vanishing coefficients are a_{3k} for k = 0, 1, 2, …

PREREQUISITES
  • Understanding of ordinary differential equations (ODEs)
  • Familiarity with power series and their convergence
  • Knowledge of recursion and series manipulation
  • Basic calculus, particularly differentiation and integration
NEXT STEPS
  • Study the method of power series solutions for ODEs
  • Learn about non-recursive formulas in series expansions
  • Explore the implications of arbitrary constants in differential equations
  • Investigate the role of coefficients in power series convergence
USEFUL FOR

Students and researchers in mathematics, particularly those focusing on differential equations, series solutions, and mathematical analysis. This discussion is beneficial for anyone seeking to deepen their understanding of ODEs and power series techniques.

Mangoes
Messages
92
Reaction score
1

Homework Statement



Solve for [tex]y' = x^2y[/tex]

The Attempt at a Solution



There's something that's been really bothering me about this question and similar ones.

We assume that the solution to the ODE will take the form

[tex]y = \sum_{n=0}{a_nx^n}[/tex]

After finding y', plugging in the expressions into the ODE, and distributing x2,

[tex]\sum_{n=1}na_nx^{n-1} - \sum_{n=0}a_nx^{n+2} = 0[/tex]

I want to eventually combine the summations, so first I make the exponents match.

[tex]\sum_{n=0}(n+1)a_{n+1}x^n - \sum_{n=2}a_{n-2}x^n = 0[/tex]

I need to make the indices match without manipulating exponents, so I just peel off the first two terms (n = 0 and n = 1) from the first summation. After doing so and combining the summations:

[tex]\sum_{n=2}((n+1)(a_{n+1} - a_{n-2}))x^n + a_1 + 2a_3x = 0[/tex]

Now here's where I start becoming uncertain...

The summation begins at n = 2, so the smallest power of x that can come out is x2. Since the RHS is 0,

[tex]a_1 = 0[/tex]
[tex]2a_3x = 0[/tex]

This means that the coefficients a_1 and a_3 must be equal to 0. Also, since the coefficients for all powers of x on the RHS are zero,

[tex](n+1)a_{n+1} - a_{n-2} = 0[/tex]

Rearranging gives me the recursion formula for n ≥ 2:

[tex]a_{n+1} = \frac{c_{n-2}}{n+1}[/tex]

My end goal in here is to write the sum in a way that doesn't involve any past terms so that there's no recursion. I start plugging in numbers and hope that I see some pattern...

n = 2, [itex]a_3 = \frac{a_0}{3}[/itex]

n = 3, [itex]a_4 = \frac{a_1}{5} = 0[/itex]

n = 4, [itex]a_5 = \frac{a_2}{6}[/itex]

n = 5, [itex]a_6 = \frac{a_3}{7} = 0[/itex]

n = 6, [itex]a_7 = \frac{a_4}{8} = \frac{a_1}{5*8}[/itex]

At this point it's clear that the odd n will give a zero and that I only have to worry about even n.

What bothers me is the term I get when n = 4. I'm not given any information as far as I know about a2 and since the ODE is first-order, there's only one arbitrary constant which I assume is a0 (if I don't assume it's arbitrary then I have no idea what to say about a0 either). So I'm stumped. How do I write a non-recursive formula for the sum with the mystery number a2? What am I missing in here?

Apologies for the lengthy post

EDIT: Disregard the thing I said about odd and even n. n = 7 doesn't equal 0.
 
Last edited:
Physics news on Phys.org
Mangoes said:
The summation begins at n = 2, so the smallest power of x that can come out is x2. Since the RHS is 0,

[tex]a_1 = 0[/tex]
[tex]2a_3x = 0[/tex]

This means that the coefficients a_1 and a_3 must be equal to 0. Also, since the coefficients for all powers of x on the RHS are zero,

I guess there's a typo there. You should have a1 = a2 = 0.
 
Because of the error clamtrox pointed out, you had a3=0, and the n=2 equation then tells you a0=0. So a2 would be the arbitrary constant.

Since you're really supposed to have ##a_1=a_2=0##, the non-vanishing coefficients are ##a_{3k}## (k=0, 1, 2, …), and a0 plays the role of the arbitrary constant.
 

Similar threads

Replies
8
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 16 ·
Replies
16
Views
2K