Goal:
This lab will help you to explore and understand some key ideas in the
study of line integrals.
You will use Maple VI
In Maple, one has access to many different libraries of specialized functions.
Among these are statistical tools in stat, graphical tools in
plots, and linear algebra tools in linalg.
We will use the latter two toolboxes.
To use the toolboxes, you must load them with the command
with(plots): with(linalg):
Background:
Consider a curve described by
for
going from
to
,
so that
is the starting point of the curve and
the
ending point.
The line integral of along the curve
with respect to
arclength
is:
One important application of the line integral is in computing work.
The work to move a particle along a curve
through a
vector force field F is given by the
line integral
If, for the same starting and ending points, no matter what path you use
in between, you always use the same amount of work to get from the
starting point to the ending point,
the vector field
is called conservative.
Otherwise, it is called nonconservative.
Maple Implementation:
We can get Maple to do a lot of work, both computationally and graphically,
for us.
Specifically, we will go through the steps to
The following example will find the work to move through the force
field
Load the plots and linalg toolboxes.
with(plots): with(linalg):
Set up the vector field and plot it.
F:=[x^ 2,y+z,z^ 2];
fieldplot3d(F,x=-1..1,y=-2..2,z=-3..3,axes=normal);
For the fieldplot3d command,
you must give the plotting function a range in
space to plot over. Also, to have the coordinate axes appear, the axes
option was included. You should use help to see more about fieldplot3d
and the accompanying options.
Use the mouse button to rotate your plot.
Now, define a curve with the parametrization
as follows:
x:=2*cos(t); y:=2*sin(t); z:=3*t;
r:=[x,y,z];
and then plot to see what it looks like.
spacecurve(r, t=0..4*Pi, axes=normal);
This curve is a helix. We need a tangent vector
to it for our integral, so we issue the command
dr:=diff(r,t);
and finally set up the integral by doing the dot product of
with the tangent vector
integrand:=innerprod(F,dr);
To get the total work, we just perform a definite integral from 0 to 4.
W:=int(integrand,t=0..4*Pi);
which should come out to be . This is the total work to move
against the given force field along the helix for two revolutions.
Note on Cleanup:
We assigned specific expressions to x,y and z. We want to
deassign those before using the commands
again, so issue the command.
x:='x'; y:='y'; z:='z';
Discussion:
The starting point of the curve above was r(0) = (2,0,0),
and the final point was r() = (2,0,12
).
You can check this by typing
t:=0; r; t:=4*Pi; r; t:='t';
Once again, it is very important that you set t back to a variable.
The total work to go from (2,0,0) to (2,0,12) along the curve given
was 576
.
Let's try another path between the two points.
Consider the curve with parametrization
for
. Does this curve connect the two points?
Check this.
Using the same set of commands we just entered, show that the total work along
this curve comes out to . (Make sure you change the limits of
integration so that you get the integral to go from
to
.)
It takes more work to go along this path than it does along the helix.
Thus, the field
is certainly
a non-conservative force field;
otherwise we would always have W=576
when we go from
to
.
Now, a little logic. Had we gotten again,
this would not necessarily mean we had a conservative
field; it would just suggest the possibility of it. There is no such thing
as ``proof by example'' but there is ``disproof by counterexample,''
which is what we just did.
As one last demonstration of why this force field above
is nonconservative, try computing the work to move along the curve as
described by
Exercise ONE
Consider the force field
A Tool for distinguishing Conservative
and Nonconservative Force Fields - CURL
At this point, we know if we find the work to go between
two points along two different paths and get different results, then we
have a nonconservative force field. If we get the same amount of work,
we only know we might have a conservative force field. Some other
path might require a different amount of work; we can't say for sure. So
we need some definite way to tell, and computing lots of line integrals
won't do it. One tool (there are several) is the curl operator.
It is easy to use so we introduce it here.
The curl operator applies only to vector fields. It gives
another vector field as the result. We will use the result below
which we will prove later on in the course.