0
$\begingroup$

I have a triple integral which is kind of complex, and I want to use Mathematics to help me do the integral. However, when I press "Enter" and "Shift" the software get stuck. I wonder whether this integral cannot be compute?enter image description here

f[x_]:=1/(A-F*x^2+x*B I);
g[y_]:=1/(A-F*y^2+y*B I);
h[z_]:=1/(A-F*z^2+z*B I);
j[x_,y_,z_]:=1/(A-F*(x+y+z)^2-(x+y+z)*B I);
result=Integrate[f[x_]*g[y_]*h[z_]*j[x_,y_,z_],x_,y_,z_]
| improve this question | | | | |
$\endgroup$
  • 1
    $\begingroup$ Your figure is illegible. Please post copyable code. $\endgroup$ – David G. Stork Mar 13 '15 at 1:39
  • $\begingroup$ you should not have the underscores in the last expression $\endgroup$ – george2079 Mar 13 '15 at 3:14
1
$\begingroup$

Your integral is too complicated for closed form. After fixing your syntax error, and integrating w.r.t. x only, you can see the result contains complex value and very complicated trig functions. Mathematica can't do the integration w.r.t. y at this stage

f[x_] := 1/(a - c*x^2 + x*b I);
g[y_] := 1/(a - c*y^2 + y*b I);
h[z_] := 1/(a - c*z^2 + z*b I);
j[x_, y_, z_] := 1/(a - c*(x + y + z)^2 - (x + y + z)*b I);
result = Integrate[f[x]*g[y]*h[z]*j[x, y, z], x]

Mathematica graphics

Integrate[result, y]  (*wait....wait...*)

Mathematica graphics

Do you think there is closed form solution for this? Have you considered numerical integration?

| improve this answer | | | | |
$\endgroup$
  • $\begingroup$ I learnt this from an article that is Non‐Normal Stochastic Response of Linear Systems. This article wants the result as the expression not the numerical value. I want to test and verify this integration whether it is supposed to be right. $\endgroup$ – whdwpy666 Mar 14 '15 at 0:48
  • $\begingroup$ I am sorry. I forgot to say the integration limits of this problem. All the upper limits are +infinity, and all the low limits are -infinity.I learnt this from an article that is Non‐Normal Stochastic Response of Linear Systems. This article wants the result as the expression not the numerical value. I want to test and verify this integration whether it is supposed to be right.@Nasser $\endgroup$ – whdwpy666 Mar 14 '15 at 0:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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