And welcome to the first lecture on Computational
Electromagnetics which is the review of Vector
Calculus. So, as we will find out in this
course the whole language in which this course
is thought or spoken is the language of Vector
Calculus. So, it is important that we become
comfortable with some basic ideas that are
useful for electromagnetics. Vector calculus
is a very vast area, we will cover only those
parts which are relevant to electromagnetics.
So, these are the topics that we will cover
in this module. We will start with the Chain
Rule of Differentiation, introduce the idea
of the gradient, move to the three most popular
operators that we will find the gradient,
divergence and the curl. There are some very
important theorems in vector calculus that
we will talk about and then some of the corollaries
which are like the tools and techniques used
in this area.
So, starting with the Chain rule; so, let
us start from the beginning all right. So,
let us consider scalar function. Scalar function
and let us say it is of three variables. It
could be of any number of variables, but we
will stick with three because where even a
three dimensional space it is easy to visualize.
So, something simple could be for example,
here is a charge plus q and I am standing
over here and there is some distance over
here, there is a vector r that connects the
source to where I am, and this observer here
could be recording; let say the electrostatic
potential right.
Now, we all know that you from you know high
school that potential is simply given by a
simple expression minus q by 4 pi epsilon
mod r where mod r is square root of x square
plus y square plus z square. So, now, you
have this function f over here which is function
of three variables x, y and z.
Now, let say that this observer wants to change
in some wants to move in some direction and
this direction could be arbitrary and I want
to know how does the potential change. So,
in other words what is this delta f; if there
is some arbitrary change in the direction
of the observer ok. So, in other words to
make it precise so, let say x change from
x to x plus d x similarly y will go to y plus
d y and z goes from z to z plus d z, we want
to keep it as general as possible.
So, again this is something that we studied
in calculus high school so, called chain rule.
So, chain rule tells us how what is the total
change in a function given changes in the
composite sort of variables. So, this df will
be given by a set a summation of the changes
due to a change in x due to change in y and
a change in z right. So, this hopefully is
familiar to all of you chain rule over here.
Note that these are partial derivatives, because
the function depends on all three variables
this only text the derivative with respect
to one of the variables. Now, when I look
at this expression while this is correct there
is a more convenient way of looking at the
same thing. So, if I look at this expression
over here, I can think of it as the dot product
between two vectors right. So, I can say that
this is let say the dot product between two
vectors v dot u and that for example, v could
be this vector and u could be this vector
right.
So, what I have got is this change over here
d f is the dot product of two vectors and
this vector over here is something which appears
time in again. So, this is defined as the
gradient of f. So, it is a vector and so,
it has three components over here and this
is the shorthand notation for it this over
here is the displacement right. So, this is
the observer’s displacement I can as well
write this as something like dl.
Usually, in this course I will not be putting
a arrow on top of the gradient of f, but it
is understood that it is a vector, another
thing I want to point out is that this is
a shorthand notation. So, I have written it
in brackets with three component; that means,
that it is a vector with three components.
I can as well write it as like this gradient
of f is equal to x hat, y hat and z hat. You
can see that this takes more time to write
then this; so, I am going to use this more
often or not it is understood when you see
something with in brackets with commas it
is a vector where I am talking about and its
components.
So, this was about the chain rule which gave
us the a very convenient way to arrive at
the idea of a gradient so, this is the gradient.
Moving on, now let us try to work with this
gradient that we have defined. So, from the
previous slide we have calculated this total
derivative d f and wrote it as a dot product
between the gradient over here and the displacement
vector over here. Now, once you have written
a differential element as it is called the
next objective would be to find out what is
the total change; let us say when I go from
some point “a” vector to some point over
here “b” vector. So, for example, I could
go like this. So, this is the integral that
I want to calculate. So, for example, this
“df” could represent a differential or
a small amount of work done and I want to
find out the total work done so, I have to
integrate along the path.
Notice that the limits of the integration
are simply a and b, the root is not being
specified ok. So, it is not this is not the
only root possible I could for example, have
gone like this or any complicated root let
I could have gone like this, does not matter
it is not specified over here. So, what you
can see is that when I start integrating this
d f right. So, I know that if I just write
integral of d f from a to b right; if you
did not know anything else, you would simply
write this as the value of the function at
in this way fb minus fa right.
But now we also have a nice identity for d
f which is in terms of the gradient right.
So, this is the relation that we arrive at
and the remarkable thing of course is that
this expression over here is path independent.
It did not matter which path I took because
this is a complete differential over here,
the result depends only on the final point
and the initial point.
So, you can see immediately a very straightforward
corollary of this is that if I am going from
one point back to the same point right. So,
in other words, the final and the starting
point are the same, then this is true right
the integral over close paths. So, when I
put on the integral sign a loop over here
like this it means that it is a integral over
a closed path starting point initial point
is the same and this is going to be zero right.
Some of you may have encountered such expressions
earlier. So, these kinds of forces if a forces
of the type grad phi we call this a conservative
force right
So, for example, we know that the Electrostatic
force is a conservative field. So, it does
no work over a close path not all forces are
conservative; so, this is a special case.
So, we have got some idea of how to work with
the gradient now.