|
|
Archived: 07 Aug 2014, 09:49
|
Author |
Message |
Shaun Culver
Site Admin
Joined: 25 Feb 2008, 13:32 Posts: 106 Location: Cape Town, South Africa
|
 Test
Equation: 
_________________ “Our imagination is stretched to the utmost, not, as in fiction, to imagine things which are not really there, but just to comprehend those things which are there.”
|
05 Mar 2008, 17:55 |
|
 |
Sameed Zahoor
Joined: 12 Mar 2008, 10:57 Posts: 69 Location: India
|
 Re: Test
I assume that 'psi' is an explicit and differentiable function of 'x'.For my convenience I choose to replace 'psi' with 'y'.Then the given integral can be written as: I = intgrl{exp(-yx)} wrt x from 0 to infinity.
An elementary use of the chain rule leads to the following expression:
I = {-1/(y+xy')}{exp(-yx)} wrt x from 0 to infinity(y' denotes the first derivative of y wrt x).To calculate the value of the integral explicitly we require an expression y(x) in terms of x.Then the value is simply:
I(infinity)-I(0)
As a special case if y is a constant function we have y'= 0 and exp{-infinity}=0.We also have exp(0)=1.Hence,
I=0-(-1/y)=1/y (y is not 0)
|
12 Mar 2008, 12:13 |
|
 |
Lemaitre
Joined: 19 Mar 2008, 13:44 Posts: 9 Location: Beirut, Lebanon
|
 Re: Test
OK Sameed! Could you please show us how you apply your expression of "I" to find the value of the original integral when y = x ?We should get 1/2*root (pi), shouldn't we?
|
19 Mar 2008, 14:27 |
|
 |
Sameed Zahoor
Joined: 12 Mar 2008, 10:57 Posts: 69 Location: India
|
 Re: Test
I is defined only when y+xy' is not zero.Therefore, I(0) is not defined when y=x.However,we can evaluate this integral by switching to polar coordinates which yields the familiar sqrt{pi}/2.Thanks for pointing it out.
|
01 Apr 2008, 05:55 |
|
 |
|
|