0
$\begingroup$

I want to solve the non-linear ODE

$$\frac{d^2}{dx^2}y=a(y+y^3)$$

With the boundary conditions that $$\lim_{x\to \pm \infty} y(x) =0$$

I am not aware of any analytical method for solving this kind of problem. So, I want to solve it numerically. Which method would be suitable for this ODE? I am aware of different methods for finite boundary conditions.

Also, does anybody know if it has an analytical solution to which I could compare my numerical solution?

$\endgroup$
  • $\begingroup$ To me this looks more like a 2nd order (partial) differential equation in one spatial variable. $\endgroup$ – Jan Jul 24 at 11:44
  • $\begingroup$ The book, "Exact Solutions for Ordinary Differential Equations" by Chapman & Hall/CRC does not seem to have a solution for this problem. Have you tried using the Runge-Kutta method. Not sure it will work for this problem, but what I remember from college, it seems like it should work for a numerical solution. $\endgroup$ – Jesse Jul 25 at 1:22
  • $\begingroup$ On further searching I found that it has an analytical solution of the form $\sqrt{\coth^2(\sqrt{a} x )-1 } $ which has singularity at the origin so no numerical method will work for it but the singularity is of the order $1/x$ so I made the substitution $ u=x^2 y $ which eliminates the singularity at the origin.In the new equation numerical methods apply. But now I am stuck I don't know which method is used for odes with infinite boundary conditions. The only idea that I have is that I make the solution close to 0 at the end points but I am not aware of any numerical for solving this nonl. $\endgroup$ – ben tenyson Jul 25 at 1:40
  • 1
    $\begingroup$ Wouldn't $y(x)=0$ satisfy this? $\endgroup$ – fibonatic Jul 25 at 6:03
  • $\begingroup$ I need a non trivial solution $\endgroup$ – ben tenyson Jul 25 at 9:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.