User Guide

Why Use Optimization? 2
-3
Wh
y
Use Optimization?
To solve the problem manually, you could assign an arbitrary
value to R1, manually calculate the current, then make an
educated guess to adjust the values until a satisfactory solution
is found. Or, you might use a simulation to sweep the value for
R1 with a DC Sweep analysis, carefully analyzing the results to
find the best solution.
These manual methods have two major disadvantages:
• Because the diode is a non-linear device, manual
calculations can be time-consuming.
• Sweeping a parameterized value can take a large number of
simulations, depending on the range and increment selected.
The PSpice Optimizer automates these processes by handling
calculations for you and intelligently directing the series of
simulations. Given results of the previous simulations, the
optimizer automatically adjusts the parameterized value of R1
for the next run, thus eliminating unnecessary iterations, which
in turn, provides a solution more quickly and with less effort.
Once the PSpice Optimizer settles on the best solution, you can
still explore available tradeoffs. When done manually, this
iterative process can be difficult and frustrating. With the
optimizer, you can tweak the parameter(s) and immediately
determine whether the design still meets specifications. You can
also change the value of the specification(s) and immediately
determine how parameter values change. If you are dissatisfied
with the result after any change, you can always return to the last
set of values.
When solvin
g
complex
problems, the manual approach
can be too unwieldy to consider.
For example:
• When your desi
g
n has
multiple parameters or
complicated parameter
interactions, you may find it’s
nearly impossible to know
which parameters to chan
g
e,
and how best to chan
g
e
them.
• When solvin
g
for multiple
specifications, the solution
often depends on the order in
which
g
oals and constraints
are optimized. This
sequential approach can
miss possible solutions since
it is impractical to repeat the
process startin
g
with a
different
g
oal or constraint
each time.