Nonlinear constraints with solvers.cp

[Richard W Lincoln]

Hello,
I have had some success using CVXOPT with my project (http://
github.com/rwl/pylon) to solve certain traditional power system
analysis problems. I am trying now to determine if I can implement an
AC Optimal Power Flow routine using the cvxopt.solvers.cp function.
This has been achieved in MATPOWER using MATLAB's fmincon function,
but I am confused as to cp's handling of nonlinear constraints. The
online user's guide mentions the evaluation of nonlinear constraints,
but the docstring in cvxprog.py talks only of linear constraints. Is
it possible to solve a problem with a nonlinear objective function and
nonlinear constraints using cp?
Regards,
Richard

[Joachim Dahl]

yes, if both the objective and the constraints are convex and twice
differentiable
then you can use cvxopt.solvers.cp().

That routine is described in §8.1:
http://abel.ee.ucla.edu/cvxopt/documentation/users-guide/node49.html

or in the docstrings,
$ pydoc cvxopt.solvers.cp

I saw your project awhile ago, googling for applications of cvxopt - it
looks
great!

Please let us know if you have more problems.
best regards
joachim

[Olalekan Ogunmolu]

Can one specify which algorithm to use for the minimization problem?

Specifically, I am looking for the interior point method of solvers.