Example
6. Use Powell's Method
to find
and
for
the function
. Use
the initial point
.
Solution 6.
![[Graphics:../Images/PowellMethodMod_gr_142.gif]](../Images/PowellMethodMod_gr_142.gif)
![[Graphics:../Images/PowellMethodMod_gr_145.gif]](../Images/PowellMethodMod_gr_145.gif)
![[Graphics:../Images/PowellMethodMod_gr_148.gif]](../Images/PowellMethodMod_gr_148.gif)
Enter the initial
point
and
set
.
For k=1 the
construct the function
.
Find
where
is
a minimum.
And construct
.
![[Graphics:../Images/PowellMethodMod_gr_160.gif]](../Images/PowellMethodMod_gr_160.gif)
For k=2 the
construct the function
.
Find
where
is
a minimum.
And construct
.
![[Graphics:../Images/PowellMethodMod_gr_166.gif]](../Images/PowellMethodMod_gr_166.gif)
Set
and
Find
where
is
a minimum.
And construct
![[Graphics:../Images/PowellMethodMod_gr_175.gif]](../Images/PowellMethodMod_gr_175.gif)
Set
.
If the minimization is done with an application of the golden ratio
or Fibonacci search then slightly different results might be
obtained.
Recall that we used Mathematica's built in procedure
FindMinimum to perform the line
searches in the above computations.
Perform the iteration using the subroutine Powell.
Let us compare this answer with Mathematica's built in procedure FindMinimum.
(c) John H. Mathews 2004