Author 
Message 
Daniel Nierro science forum beginner
Joined: 19 Jul 2006
Posts: 2

Posted: Wed Jul 19, 2006 10:28 am Post subject:
Vector field flow problem  help?



Hi!
I'm struggling with some calculations of what the flow of a vector field
through a surface is.
The vector field is F(x,y,z) = (e^(xyz) + y, e^(xyz)  z, x)
and the surface (S) is the part of the paraboloid z = x^2 + y^2 where z <=
x  y + 2.
(The surface S is oriented so it's normal points away from the zaxis.)
I think I know how to attack this problem; by completing the surface S with
a "lid" (L) so it forms a closed object (O), and then the value of the
sought value of the surface integral becomes equal to the triple integral
over the solid closed O object minus the surface integral over the lid L.
I think the proper lid L is z = x  y + 2, but then I get kind of stuck in
my calculations after a while.
Am I attacking this in the wrong way?
I've been struggeling with this for quite some time now and becoming a bit
desperate, does anyone know what the solution of this is?
Very grateful for any feedback,
Daniel Nierro 

Back to top 


Lynn Kurtz science forum Guru
Joined: 02 May 2005
Posts: 603

Posted: Wed Jul 19, 2006 6:12 pm Post subject:
Re: Vector field flow problem  help?



On Wed, 19 Jul 2006 12:28:45 +0200, "Daniel Nierro"
<dani4965@student.uu.se> wrote:
Quote:  Hi!
I'm struggling with some calculations of what the flow of a vector field
through a surface is.
The vector field is F(x,y,z) = (e^(xyz) + y, e^(xyz)  z, x)
and the surface (S) is the part of the paraboloid z = x^2 + y^2 where z <=
x  y + 2.
(The surface S is oriented so it's normal points away from the zaxis.)
I think I know how to attack this problem; by completing the surface S with
a "lid" (L) so it forms a closed object (O), and then the value of the
sought value of the surface integral becomes equal to the triple integral
over the solid closed O object minus the surface integral over the lid L.
I think the proper lid L is z = x  y + 2, but then I get kind of stuck in
my calculations after a while.
Am I attacking this in the wrong way?
I've been struggeling with this for quite some time now and becoming a bit
desperate, does anyone know what the solution of this is?
Very grateful for any feedback,
Daniel Nierro

Let's use n for the unit normal directed outwards from the volume. So
you want the surface integral Int_S F dot n dS and you have observed
(Int_S + Int_L) F dot dS = Int_O del dot F dV
You have probably noticed that del dot F is zero so if you can
calculate the integral over the lid: Int_L F dot dS, you are home
free. So far, so good, and no, I don't think you are attacking the
problem the wrong way.
Expressing your lid parametrically using x and y as the parameters:
R(x,y) = < x, y, x  y + 2 >
R_x = < 1, 0 1 > and R_y = < 0, 1, 1 >
R_x cross R_y = < 1, 1, 1 >
so your ndS vector becomes < 1, 1, 1 > dy dx.
The surface integral over your lid is:
int_L < exp( x  y  z) + y, exp( x  y  z)  z, x > dot < 1, 1, 1 >
dy dx
What you need to do now is substitute z = x  y + 2 to get everything
in terms of x and y, simplify, and get the appropriate limits for the
intersection of the plane and the paraboloid. I haven't worked the
rest of the details out. You may find a switch to polar coordinates
helpful.
Lynn 

Back to top 


Google


Back to top 



The time now is Sun Nov 18, 2018 12:29 am  All times are GMT

