Integration with Trigonometric Substitution

Click For Summary

Discussion Overview

The discussion revolves around the evaluation of a specific integral involving a trigonometric substitution, particularly focusing on the integral of the form [(a^2 - s^2)^1/2]/(x-s) ds over the bounds from -a to a. Participants explore the implications of the integrand's structure on the choice of integration techniques, including potential singularities and the use of the Cauchy Principal Value.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant expresses difficulty with the integral due to the presence of (x-s) in the denominator, which disrupts the symmetry typically leveraged in trigonometric substitution.
  • Another participant warns about potential issues when |x| < a, indicating a need for caution in evaluating the integral under certain conditions.
  • A later reply suggests that the integral may be evaluated using the Cauchy Principal Value and the Residue Theorem, while also mentioning that trigonometric substitutions could still be applicable, though perhaps not necessary.
  • One participant introduces the Weierstrass Substitution Method as a technique for dealing with rational trigonometric integrands, proposing it as a way to convert the integrand into a more manageable form.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the best approach to evaluate the integral. There are multiple competing views regarding the applicability of trigonometric substitution and the implications of singularities in the integral.

Contextual Notes

Participants note the presence of a singularity in the integral and discuss the potential need for advanced techniques like the Residue Theorem, indicating that the discussion may involve complex analysis concepts.

dollarbills10
Messages
3
Reaction score
0
Hi,

I am seeking some input for an integral I have been stumped on for a few days.

This is the integral:

[(a^2 - s^2)^1/2]/(x-s) ds evaluated over the bounds from -a to a. The symmetry of the integration area allows the integral to be evaluated from 0 to a, and doubled.

I have always been conditioned to use trig substitution for an integral when an expression such as (or usually, exactly as) (a^2 - s^2)^1/2 is in the integrand. However, this is unique in that this expression is being divided by x-s in the integrand.

Any suggestions?

Thank you in advance, I am new so I apologize if I have not presented my topic in the most ideal format.
 
Physics news on Phys.org
The term (x-s) destroys the symmetry.
 
Hi,

Mathman, thank you for your comment.

I have attached a much better pictorial of the integrand in discussion. I have failed in trying to decipher the latex language in the forum.

I have also included my trig substitution. However, as I have hinted, I have no clue as to where to go from here.

Any comments appreciated.
 

Attachments

  • eqnshw2.JPG
    eqnshw2.JPG
    3.6 KB · Views: 404
I have never seen this before. Be careful if |x| < a.
 
Mathman,

Thank you again for the comment. Obviously, the singularity is present.

After some more investigation, the solution will definitely follow that of Cauchy Principal Value integrals, in as much as the Residue Theorem and the poles of the integral in the upper half of a complex plane are used to evaluate the integral. However, I believe you can still use trig substitutions, but it may not be necessary. If interested, I'll keep updating.

Thanks.
 
dollarbills10 said:
However, as I have hinted, I have no clue as to where to go from here.
Any comments appreciated.

Whenever you have a rational trigonometric integrand, The Wieirstrass Substitution Method [also called the Tangent Half Angle Method] is a powerful technique to use. It converts the rational trig functions to rational algebraic functions so that one can then use Partial Fractions or U Substitution or Long Division or Complete the Square methods.

http://en.wikipedia.org/wiki/Weierstrass_substitution

Here is a sample problem showing how it works.
Click on "Show Steps" in the upper right corner to see the details of the solution.

http://www.wolframalpha.com/input/?i=integral+[+sin+x+%2B+cos+x+]+%2F+[+1+%2B+cos+x+]+dx

You may need to cut and paste in the link

It is not as difficult as it looks once you understand what is going on between "phi" and "phi / 2"

Enjoy
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 21 ·
Replies
21
Views
4K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K