You are on page 1of 14

A Note on Integrals & Hybrid Contours in the Complex Plane

John Gill July 2014


Abstract: Contour integrals can be expressed graphically as simple vectors arising from a secondary contour.
We start with a well-behaved (continuous, analytic, etc.) function in the complex plane:
(1) ( , ) z t , with z S a convex set in C and [0,1] t . ( , ) and [0,1] z t S z S t
Let
(2)
1
, ,
( ) ( , )

= +
k
k n k n n
g z z z , with
,
( )
k n
g z S , 0 k n <
Require
1, 2, ,
0 1 < < < < =
n n n n
and
,
lim 0
k n
n

= , where 1, 2,..., k n = .
Set
1, 1,
( ) ( )
n n
G z g z = ,
( )
, , 1,
( ) ( )
k n k n k n
G z g G z

= and
,
( ) ( )
n n n
G z G z =
with ( ) lim ( )
n
n
G z G z

= , when that limit exists.


Written in an alternate iterative form
(3)
1
, 1, , 1, 0,
( , ),


= +
k
k n k n k n k n n n
z z z z ,
the distribution of points forms a Zeno contour
(4)
{ }
,
1
( ) ( ) lim ( )
=

= =
n
n k n n
k
n
z .


The word Zeno denotes the infinite number of actions required in a finite time period if
, k n

describes a partition of the time interval [0,1].



( , ) z t is associated with a unique time-dependent vector field (TDVF):

(5) : ( , ) F F z t , where ( , ) ( , ) = + F z t z t z .

Under benign conditions (3) admits an equivalent closed form:

(6) ( ) : ( ), ( , ), (0) = = =
dz
z z t z t z
dt
.

Now suppose another well-behaved function ( , ) f z t S is introduced. Set
(7)
*
( , ) ( , ) ( , ) z t f z t z t = and create a new, hybrid contour in the following way:
(8)
* * *
1
, 1, , 1, 0,
( , ),
k
k n k n k n k n n n
z z z z


= + ,
with
{ }
* *
,
( )
n k n
z = and
* *
( ) ( ) as
n
n

Observe that
*
is a function of points on the original Zeno contour . So that we now have two
contours: ( ) (in green) and
*
( ) (in red) that are siamese, i.e., originating at the same
point. The underlying TDVF is illustrated by vector clusters (black for t=0, green for t=1):




Example 1:





Rewrite (3) as
1 1 1
1, , 1, 1, 1,
1 1
1
( , ) ( ) ( , ) ( , )
n n
k k k
k n k n k n k n k n n n n
k k
f z z z f z z
n



= =
=


(9)
* *
1
1,
1
( )
1
lim ( , ) ( )
n
k
k n n
n
k
fdz z G
n

=
(
= + =
(


Thus the value of the integral is essentially the vector connecting to
*
( ) G .
In Example 1 above
2
2 2( 2 )
1 2 1 2
x y t
i
t t


= +
+ +
, f = and 1 .5i = + . Then
( )
1.6627 4.0006 dz i

+






If f = , a very well-behaved function,
(10)
0
1 1
2
( ) 0 0
( ) ( ) ( ( ))
z
dz
z dz z dt z t dt
dt

= =

or
0
(1)
( )
z
z
z dz

for analytic .
The notation
( )
fdz

represents
( )
( ) f z dz

or
1
0
( ( ), ) ( ( ), ) f z t t z t t dt





Example 2: ( ) ( ) xCos yt iySin xt = + , (2 ) ( ) f x t i x y t = + + , 1 i =

( )
5.4003 1.6223 fdz i




Example 3: A simple case where direct evaluation is easily done:
( ) 2 z z =
2
( )
t
z t e = , ( , ) f z t zt = , .5(1 ) i = +
( )
10.2995 fdz i





Example 4: ( ) ( ) xCos xt y ySin xt y i = + + , 2 ( ) 2 ( ) f Sin x yt Cos x yt i = + , 5(1 ) i = +

( )
3.3396 5.5439 fdz i

+










Virtual Integral vs secondary contour: Assume ( ) z = . Previously, a virtual integral was
established in the following way:

( )
n
:
, 1, 2, 1,
1 1 1 1
( ) ( ) ( ) ( )

= + + + + +
n n n n n n
z z z z
n n n n


Now, by slight of hand, define
( ) [ ] ( ) ( )
1
1,
0
, , 0,1 and , lim , with , defined

| |

|
\

mk mn
m
k
t t z t dt
n
:

1
0
1 1 1 2 1 3 1
( ) , , , , ( , )
| | | | | | | |
= + + + +
| | | |
\ \ \ \

n
n
G t dt
n n n n n n n n

So that
1
0
( , ) ( ) =

t dt G .

Under perfect circumstances ( ): ( ) z z t = has a pleasant closed form and
( , ) ( ( )) , (0) = = t z t z . For example,
2
( , ) 2 ( ( )) 2 ( , ) = = =
t
z t zt z t te t
Therefore
1
0
( ) ( ( )) ( ) = =

z t dt G , whereas, in this note,


1
2 *
0
( ( )) ( ) =

z t dt G .
However, it is usually the case that ( , ) t cannot be easily described as ( ( )) z t and
1
0
( ) ( , ) ( ) = =

t dt G has a virtual integrand.


The notation
1
2 *
0
( ( )) ( ) =

z t dt G is used even when ( ) z t is indescribable.




Example 5: ( ) ( ) xCos yt iySin xt = + , 2.5 2 = i

1
0
( ) ( , ) ( ) .1789 1.7624 = =

z t dt G i ,
and
1
2 *
0
( ( )) ( ) 1.8830 .4950 = +

z t dt G i . ( Results hold for ( , ) z t )





Simplified, heuristic description of the process:
When the functions are analytic and integrations in closed forms are possible, we have
( ) ( )
dz
z z z t
dt
= = and ( ) ( )
d
f z z
dt

= , leading to
( ( )) ( ( )) ( ) d f z t z t dt f z dz = =
1
( ) 0
(1) (0) ( ) ( ( )) ( ( )) f z dz f z t z t dt

= =


Results are valid for ( , ) f f z t = and ( , ) z t = .

Equation solving . . .
Let us restrict our discussion to f = . Then what has been described previously fits the format
( , ) T =
( )
( ) z dz

=


That is to say : ( ) T C S S C where ( ) C S is the set of complex functions continuous
on the set S and C is the complex plane.

Example 6: ( ) ( ) ( ) ( ) z x iy xCos xy iySin xy = + = + , with associated vector field
: ( ) ( ) F z z z = + F . Thus

( 2.5 1.8 )
( ( ) ( ), 2.5 1.8 ) ( ) 1.038 .3289
i
xCos xy iySin xy i z dz i

+
+ + =






Example 7: Determine satisfying the equation ( ) 2 , 2i z = . From 2
dz
z
dt
= , one finds
2 2
( ) (1)
t
z t e z e = = . Therefore
2
2 2 .1366(1 )
e
zdz i i

= +



Example 8: Given
2
( ) (1 2 ) 2 , 1 1.2 z t t t i i = + + = , find .
From ( , )
dz
z t
dt
= we see that
2 4 (1 )
2 4 ( , )
1 2 1 2
dz z t t
ti i z t
dt t t

+
| |
= + = + =
|
+ +
\
.
Hence
3 2 2 (1 )
: ( ) 2
1 2 1 2
t y t t
f z x i y
t t
+ + +
| | | |
= + +
| |
+ +
\ \
F , a TDVF
And ( )
1 1
2
2
0 0
( ( )) 4 2 2.5046 1.5950 z t dt ti dt i = +







Example 9:
1
z
= and 2 i = + . Solve for : ( , ) = . From
1 dz
dt z
= ,
2
( ) 2 z t t = + and the problem looks like this:
2
2
1
2 dz i
z

+
= +

. Thus
2 2
2
i
e
+
+ =
gives a solution: .10168 .16128i .






Attempting to solve ( , ) = for requires more effort and may entail restricting the
functional form.


Example 10: Solve for C :
( , ) 1 2 Cz i i = +
From
dz
Cz
dt
= one obtains ( )
Ct
z t ie = . Therefore 1 2
C
ie
i
C z dz i = +

. Solving for C requires


numerical techniques for intrinsic functions.
.2724 3.9117 C + .






Example 11: Solve for C : ( ,1) 2 3 iCz i = + 6 4 C i = +


Example 12: Given
2
( ) ( 1) 2 z t C t ti = + and 1 = , solve ( ,1) 4 2 T i = for .
2( ) C t i = and numerical computations give 1.9734 1.8800 C i +





Final Comments . . .
Setting
1
0
( ) ( , ) t dt =

and
1
( )
( ( )) f G

= , it is possible to write
( ) ( )
( )
( ) ( ) ( ) '( ) 1 ( ) ( ) f z zf z dz f

=


Given a primary and a secondary contour,
( ) :
1
, 1, , 1,
( , )
k
k n k n k n k n n
z z z


= + and
*
( ) :
* *
1
, 1, , 1,
( , )
k
k n k n k n k n n
z z T z


= +

*
( )
( )
( )
( )
T z
dz G
z



Enough of this trivia. Im getting bored!