2
$\begingroup$

Context

I would like to show that these 2 functions are identical

f1[a_] = (2/π) Integrate[Cos[t]/(1 -2a Cos[t] + a^2)^(3/2), {t, 0,π}, 
    Assumptions -> {0 < a < 1}] // FullSimplify

enter image description here

corresponding to this integral

$$ \frac{2}{\pi } \int_0^{\pi } \frac{\cos (\theta )}{\left(1-2 a \cos (\theta )+a^2\right)^{3/2}} \, d\theta $$

and

f2[a_] = (4/(π a (1 - a^2)^2))((1 + a^2) *
           EllipticE[a^2]-(1 - a^2) EllipticK[a^2])

enter image description here

This does not seem to work:

f2[a]/f1[a]  // FullSimplify[#, Assumptions -> {0 < a < 1}] &

Any suggestions?

$\endgroup$
3
  • 1
    $\begingroup$ Some evidence, but not complete solution: Series[f2[a] - f1[a], {a, 0, 100}, Assumptions -> {0 < a < 1}] and # == 0 & /@ (f2[a] - f1[a] /. a -> RandomReal[1, 100, WorkingPrecision -> 20]) // Apply[And] $\endgroup$
    – Michael E2
    Commented Oct 24, 2020 at 16:01
  • 1
    $\begingroup$ It turns out that MA can do a lot with elliptic integrals, however, it does not know some identities, in particular Eq.8.126 from Table Of Integrals, Series And Products by Gradshteyn and Ryzhik. I retype them below in original notations $K\!\left(\frac{2\sqrt{k}}{1+k}\right)=(1+k)K(k),$ and $E\!\left(\frac{2\sqrt{k}}{1+k}\right)=\frac1{1+k}\left(2E(k)-(1-k^2)K(k)\right).$ $\endgroup$
    – yarchik
    Commented Oct 25, 2020 at 6:43
  • 1
    $\begingroup$ See also comments to the answer to this post mathematica.stackexchange.com/questions/204969/… $\endgroup$
    – yarchik
    Commented Oct 25, 2020 at 6:43

1 Answer 1

7
$\begingroup$

You can use the substitution rules based on the Eq.8.126 from Table Of Integrals, Series And Products by Gradshteyn and Ryzhik. They are called quadratic arithmetic-geometric mean (AGM) transformations for the complete elliptic integrals. I retype them below in original notations $$K\!\left(\frac{2\sqrt{k}}{1+k}\right)=(1+k)K(k),$$ and $$E\!\left(\frac{2\sqrt{k}}{1+k}\right)=\frac1{1+k}\left(2E(k)-(1-k^2)K(k)\right).$$ One should remember that they are valid for $0\le k\le 1$ and one should beware of slightly different MA conventions.

rules={EllipticK[(4 λ_)/(1+λ_)^2]-> (1+λ)EllipticK[λ^2], 
EllipticE[(4 λ_)/(1+λ_)^2]-> 1/(1+λ) (2EllipticE[λ^2]-(1-λ^2)EllipticK[λ^2])};

Now with

f1[a_] = (2/π) Integrate[Cos[t]/(1 - 2 a Cos[t] + a^2)^(3/2), {t, 0, π}, 
Assumptions -> {0 < a < 1}];
f2[a_] = (4/(π a (1 - a^2)^2)) ((1 + a^2) EllipticE[a^2] -
        (1 - a^2)EllipticK[a^2]);

Then

Simplify[(f1[a] /. rules) - f2[a]]

returns

(* 0 *)

$\endgroup$
2
  • $\begingroup$ @chris Thank you for making my post more clear. $\endgroup$
    – yarchik
    Commented Oct 25, 2020 at 12:14
  • $\begingroup$ Thank you for your work! $\endgroup$
    – chris
    Commented Oct 25, 2020 at 12:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.