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

Should I expand as a power series and then integrate or is there an elegant method?

  1. Apr 28, 2008 #1
    [tex] \int_{0}^{\pi/2}(cos(\theta)*exp(-2[\pi(1-cos(\theta))]^2)/k^2)/erf(2\pi/k) [/tex]

    Where of course the error function erf is defined as:

    Anyway... this is the problem I want to integrate. I am not looking for someone to post a solution. My question is simply what is the best way of tackling this monster. My first thought is to expand each of the functions into a power series and use the "crank the handle" method. Not very elegant. Can anyone see a better/quicker method?


  2. jcsd
  3. Apr 28, 2008 #2
    \int_{0}^{\pi \over 2} cos(\theta) \frac{\exp\left ( \frac{-2\pi(1-cos\theta)^2}{k^2}\right ) }{erf(\frac{2\pi}{k})}d\theta

    Now, since erf doesn't depend on theta, you can take it out of the integral. After that, your best bet is to try series. It's possible you might be able to do this with contour integrals, although that might be equally as difficult, if even possible.
  4. Apr 30, 2008 #3
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook