Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Nonlinear Differential Equation solving help please!

  1. May 17, 2012 #1
    1. The problem statement, all variables and given/known data
    Consider the following differential equation:
    State whether this equation is linear or nonlinear and find all it's solutions

    2. Relevant equations
    I think that the Bernoulli differential equation is relevant, but I'm not sure:

    3. The attempt at a solution
    Ok, so this is really a past exam question, and I've been struggling to remember the method & can't find it anywhere in my notes.

    First, I tried to rearrange into something similar to the Bernoulli equation, so I could solve using the method from the wikipedia article (wikipedia dot org slash)wiki/Bernoulli_differential_equation.

    The rearrangement I got is


    and then I'd use the substitution w=1/y, w'=(-1/y^2)y' but I'm not sure what to do about that pesky -1/y2 in the middle. If someone could point me towards the right method to sovle this, I'd be very greatful.

    N.B. The course I'm on doesn't cover nonlinear DE's very much, and the only ones that we do have exact solutions. So this should have an exact solution.
    Last edited: May 17, 2012
  2. jcsd
  3. May 17, 2012 #2
    Edit. Yes you can use Bernoulli's Equation since you have the y^2

    Last edited: May 17, 2012
  4. May 17, 2012 #3

    I might be neglecting something that makes this more difficult, but wouldn't a basic "z(x)=y/x" substitution work?

    I mean, if you divide the original equation with x², you get

    [itex]\frac{dy}{dx} = 1 - \frac{y}{x} + \frac{y^2}{x^2}[/itex]

    I haven't studied DEs that much and have never had to use anything fancy like the Bernoulli differential equation, but from what I can tell, z=y/x seems obvious here. The DE that follows after the substitution is easy to solve, unless I made a mistake somewhere.
  5. May 17, 2012 #4
    That's what I thought, but dividing


    by y2 doesn't yield something of the form Q(x), it leaves


    Unless I'm approaching that method entirely wrong, which I may well be, I don't see how I can use the [itex]w'=\frac{-1}{y^{2}}y'[/itex] substitution to then solve it.
    Last edited: May 17, 2012
  6. May 17, 2012 #5
    I am sorry you have a Riccatis Equation.

    Take a look at how to solve this equation and then give it a try.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook