## Trying to get exact solution for a nonlinear DE

Ok, this should be quite simple. I've been looking at this problem for quite some time now, and I'm tired.. Please help me!

The equation to solve is $r'$ = $r(1-r^2)$

The obvious thing to do is to do partial fraction expansion and integrate(from r$_{0}$ to r):

$∫ (\frac{1}{r} + \frac{1}{1-r^2})dr = t$

After some trig substitution: $ln[r\sqrt{\frac{r+1}{1-r}}] = t$ (evaluated from r$_{0}$ to r). Then take the exponential on both sides.

$r\sqrt{\frac{r+1}{1-r}} = r_{0}\sqrt{\frac{r_{0}+1}{1-r_{0}}}e^{t}$

This leads to kind of a nasty expression which can't be solved explicitly for $r(t)$ (at least it seems that it can't be solved explicitly).. So this is where I'm stuck.

The solution to this ploblem is: $r(t) = \frac{r_{0}e^{t}}{\sqrt{1+r_{0}(e^{2t}-1)}}$

 PhysOrg.com science news on PhysOrg.com >> Galaxies fed by funnels of fuel>> The better to see you with: Scientists build record-setting metamaterial flat lens>> Google eyes emerging markets networks
 $∫ (\frac{1}{r} + \frac{1}{1-r^2})dr = t$ is false. r is missing on the second fraction.
 Thank you! Got it now :)

Recognitions:
Gold Member
Staff Emeritus

## Trying to get exact solution for a nonlinear DE

 Quote by Paalfaal Ok, this should be quite simple. I've been looking at this problem for quite some time now, and I'm tired.. Please help me! The equation to solve is $r'$ = $r(1-r^2)$ The obvious thing to do is to do partial fraction expansion and integrate(from r$_{0}$ to r): $∫ (\frac{1}{r} + \frac{1}{1-r^2})dr = t$
That's an incorrect "partial fraction expansion". $1- x^2= (1- x)(1+ x)$ so
$$\frac{1}{r(1- r^2)}= \frac{1}{r}- \frac{1}{r- 1}- \frac{1}{r+ 1}$$
There is no "trig substitution" involved in integrating that. the integral is
$$ln(r)- ln(r-1)- ln(r+1)+ C= ln\left(\frac{r}{r^2-1}\right)+ C$$

 After some trig substitution: $ln[r\sqrt{\frac{r+1}{1-r}}] = t$ (evaluated from r$_{0}$ to r). Then take the exponential on both sides. $r\sqrt{\frac{r+1}{1-r}} = r_{0}\sqrt{\frac{r_{0}+1}{1-r_{0}}}e^{t}$ This leads to kind of a nasty expression which can't be solved explicitly for $r(t)$ (at least it seems that it can't be solved explicitly).. So this is where I'm stuck. The solution to this ploblem is: $r(t) = \frac{r_{0}e^{t}}{\sqrt{1+r_{0}(e^{2t}-1)}}$ Again, this should be fairly simple, but I'm stuck.. Please help me!
You have integrated incorrectly.