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

I Help needed; problematic integral

  1. Feb 6, 2017 #1
    Hello.

    I am having a lot of trouble trying to solve/analyse this integral:

    $$\displaystyle \int_2^\infty \frac{x+y}{(y)(y^2-1)(\ln(x+y))} dy$$

    I have tried everything with no result; it seems impossible for me to work with that natural logarithn.

    I have also tried to compute it, as it converges for positive values of x, but that does not help neither.

    It is given that that x will be an integer x>1.

    Is there any way to leave the integral in terms of x? And to get its asymptotic behaviour?

    Thank you very much.
     
  2. jcsd
  3. Feb 6, 2017 #2

    Svein

    User Avatar
    Science Advisor

    Try splitting the integrand into three or four parts: [itex] \frac{x+y}{y(y^{2}-1)\ln(x+y)}=\frac{A}{y}+\frac{B}{y-1}+\frac{C}{y+1}+\frac{D}{\ln(x+y)}[/itex].
     
  4. Feb 6, 2017 #3
    Thank you for your interest.

    If I tried that, wouldn't I have a sum of divergent integrals?
     
  5. Feb 7, 2017 #4

    Svein

    User Avatar
    Science Advisor

    I do not know, but since the lower integral limit is 2 and x is greater than 1 you should be safe.
     
  6. Feb 7, 2017 #5
    Breaking it in 4 fractions would definitively lead to a sum of divergent integrals, but dividing it into 2 integrals:
    [tex] x \displaystyle \int_2^\infty \frac{1}{y(y^2-1)\log(x+y)} + \int_2^\infty \frac{1}{(y^2-1)\log(x+y)} [/tex]

    Since the value of the second one is too small for large x, any idea about what to do with the first one?
     
  7. Feb 8, 2017 #6

    Svein

    User Avatar
    Science Advisor

    I am speculating about the complex domain and the residue theorem, but as of now I have not figured out exactly how to apply them.
     
  8. Feb 12, 2017 #7

    Svein

    User Avatar
    Science Advisor

    I can give some sort of an answer, but not very useful.

    1. Substitute p for x and z for y (just in order not to get confused).
    2. Assume p ≥ 3
    3. The integrand now reads [itex]\frac{z+p}{z(z-1)(z+1)log(z+p)} [/itex]
    4. Calculate the residues at z = -1, 0, 1, 1-p:
    [itex]Res_{z=-1}:\frac{p-1}{(-1)(-2)\log(p-1)}=\frac{p-1}{2+\log(p-1)} [/itex]
    [itex]Res_{z=0}:\frac{p}{1(-1)\log(p)}=\frac{-p}{\log(p)} [/itex]
    [itex]Res_{z=1}:\frac{p+1}{1\cdot 2\log(p+1)}=\frac{p+1}{2 \log(p+1)} [/itex]
    [itex]Res_{z=1-p}:\frac{1}{(1-p)(2-p)(-p)}=\frac{-1}{p(1-p)(2-p)} [/itex]

    Given that, we can find the primary Cauchy value of [itex] \int_{-\infty}^{\infty}\frac{z+p}{z(z-1)(z+1)log(z+p)}dz[/itex] from the sum of the residues times πi:
    [tex] \pi i (\frac{-p}{\log(p)}+\frac{p-1}{2\log(p-1)}+\frac{p+1}{2\log(p+1)}-\frac{1}{p(1-p)(2-p)})[/tex]
    or, rewritten with x instead of p:
    [tex] \pi i (\frac{-x}{\log(x)}+\frac{x-1}{2\log(x-1)}+\frac{x+1}{2\log(x+1)}-\frac{1}{x(1-x)(2-x)})[/tex]
    Given that x should be an integer greater than 1, we have excluded the possible value of x = 2, where we get a (sort of) double pole at -1.
     
  9. Feb 16, 2017 #8
    Thank you! That helps a lot in my problem in spite of being focused on the complex domain of the function
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Help needed; problematic integral
Loading...