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

B First Order Non-Linear ODE (what method to use?)

  1. Feb 14, 2017 #1
    Hi,

    The problem is to solve:
    dy/dx = −[2x + ln(y)]*(y/x)

    Attempt:
    I have tried to see if it is exact, I found it not to be, I can't easily find a function to multiply by to make it exact either (unless I am missing something obvious). It clearly isn't seperable, nor is it homogenous (I know although some non-homogenous ODEs can be solved using the homogenous method I don't think this one can be), it also isn't a linear equation so we can't use integrating factors.

    I'm looking for pointers in the right direction/which method to use.

    Thanks for your time
     
  2. jcsd
  3. Feb 14, 2017 #2

    jedishrfu

    Staff: Mentor

    Have you tried guessing a solution?

    Something like ##a^x## or something that works well with the natural log function.
     
  4. Feb 15, 2017 #3

    jedishrfu

    Staff: Mentor

    Did you figure it out?
     
  5. Feb 17, 2017 #4
    I think you are going the right way; checking the exact form, homogeneity etc.
    Here since there is a logarithmic function, I recommend substituting y=et where t is parameter and then be solved by finding integrating factor.
     
  6. Feb 22, 2017 #5
    Using y = exp(qx), where q is a function of x , I got it to separate, giving
    q = (A/x2) - 1, where A is a constant of integration.
    It seems to check.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted