MathGroup Archive 2002

[Date Index] [Thread Index] [Author Index]

Search the Archive

RE: RE: Can you help me to solve this Integrate using Mathematica ?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg36269] RE: [mg36248] RE: [mg36221] Can you help me to solve this Integrate using Mathematica ?
  • From: "DrBob" <drbob at bigfoot.com>
  • Date: Thu, 29 Aug 2002 01:37:53 -0400 (EDT)
  • Reply-to: <drbob at bigfoot.com>
  • Sender: owner-wri-mathgroup at wolfram.com

The integral doesn't converge when b is real, because of the second
order singularity at Abs[b] in that case.  Arg[b^2]!=0 is used to
express this because the issue is whether that singularity is on the
positive x-axis.

Behavior at 0 and Infinity are fine if Re[a]>0, but otherwise those are
problems too, so Mathematica has the right answer in those terms.  I
can't vouch for the formula it comes up with when the conditions are
met, but it's not closed-form in the usual sense anyway; but it's
closed-form in terms of functions MATHEMATICA is comfortable with!

Bobby Treat

-----Original Message-----
From: David Park [mailto:djmp at earthlink.net] 
To: mathgroup at smc.vnet.net
Subject: [mg36269] [mg36248] RE: [mg36221] Can you help me to solve this Integrate
using Mathematica ?


Valdeci,

It appears that the integral will not converge if b is real. Mathematica
gives...

Integrate[Exp[-a/x]/(x^2 - b^2), {x, 0, Infinity}]
If[Re[a] > 0 && Arg[b^2] != 0, (1/2)*Sqrt[-(1/b^2)]*
   (2*CosIntegral[a*Sqrt[-(1/b^2)]]*
     Sin[a*Sqrt[-(1/b^2)]] + Cos[a*Sqrt[-(1/b^2)]]*
     (Pi - 2*SinIntegral[a*Sqrt[-(1/b^2)]])),
  Integrate[1/(E^(a/x)*(-b^2 + x^2)), {x, 0, Infinity}]]

If b is Real then Arg[b^2]==0 and Mathematica doesn't solve it. Let's
define
a function that allows us to test specific values of a and b.

f[a_, b_][x_] = Exp[-a/x]/(x^2 - b^2);

Integrate[f[2, 3][x], {x, 0, Infinity}]
Integrate::"idiv" : "Integral of 1/(E^(2/x)*(-9 + x^2)) does not
converge on
\
{0, Infinity}."

But if we use an imaginary value for b...

Integrate[f[2, 3I][x], {x, 0, Infinity}]
%//N
(1/6)*(2*CosIntegral[2/3]*Sin[2/3] +
   Cos[2/3]*(Pi - 2*SinIntegral[2/3]))
0.254022

David Park
djmp at earthlink.net
http://home.earthlink.net/~djmp/

From: Valdeci Pereira Mariano de Souza [mailto:vpms at rc.unesp.br]
To: mathgroup at smc.vnet.net

Dear Group

I would like to know, if is possible to solve

Integrate[(exp[-a/x])/(x^2-b^2),{x,0,infinity}] with

a and b constant ( Real )

using Mathematica.


Thankfully,


Valdeci Mariano


*********************************************************************
Valdeci Mariano de Souza
Master´s Degree of Applied Physics  -  Unesp/Rio Claro - State of São
Paulo - Brazil
Laboratory of Electrical Measurements
email : vpms at rc.unesp.br   or   valdecimariano at bol.com.br
phone : ( 0XX19 ) 526 - 2237
*********************************************************************









  • Prev by Date: Re: SIR epidemic Model
  • Next by Date: Re: SIR epidemic Model
  • Previous by thread: RE: Can you help me to solve this Integrate using Mathematica ?
  • Next by thread: ReadList and NumberPoint ?