I want to solve this second-order semilinear ODE $$ x''(s)-3x(s) x'(s)+x^3(s)=0, ~s\in\mathbb{R}.$$ I have tried this substitution $$p(x)=x'(s),$$ which implies $$x''(s)=p(x)\frac{d p(x)}{dx},$$ with which I reduced the original ODE to a first-order nonlinear non-homogeneous ODE. But this method is too clumsy. Is there any quick solver?

Maple has given the solution as $$x \left( s \right) ={\frac {-2\,{ C_1}\,s-2\,{C_2}}{{ C_1 }\,{s}^{2}+2\,{ C_2}\,s+2}}, $$ and I am very curious of how do Maple solve?

  • $\begingroup$ wolfram alpha has a different answer. wolframalpha.com/input/?i=y''-3yy'%2By%5E3+%3D+0 $\endgroup$ – Guy Mar 16 '14 at 8:35
  • $\begingroup$ maybe that is what you meant and don't know $\TeX$. should I edit? $\endgroup$ – Guy Mar 16 '14 at 8:35
  • $\begingroup$ Yeah it is okay now. $\endgroup$ – Guy Mar 16 '14 at 8:41
  • $\begingroup$ Sorry, I have just convert Maple's result into LaTeX directly, without considering the convention. I have modified Maple's solution. $\endgroup$ – nuage Mar 16 '14 at 8:41
  • $\begingroup$ yeah it is okay now. $\endgroup$ – Guy Mar 16 '14 at 8:41

The homogeneous EDO is successively transformed to homogeneous then separable EDO as shown below :

enter image description here

| cite | improve this answer | |
  • $\begingroup$ I get $t=1-\frac{c_1}{X}\pm\sqrt{\dots}$ $\endgroup$ – ccorn Mar 23 '14 at 2:10
  • $\begingroup$ That sign error in the quadratic formula for $t$ gets carried through to $x$. Just flip the sign of $x$ and the solution is OK. $\endgroup$ – ccorn Mar 23 '14 at 3:22

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.