Dear Dr. Ray Zimmerman,

Thank you so much for your explanation. I'll try your approach.

Kind regards,
Ni Le

On Tue, Sep 16, 2014 at 10:08 PM, Ray Zimmerman <[email protected]> wrote:

> That option is used to set the default value of the feastol parameter for
> mips(). It is really termination tolerance, not something intended to be
> used to relax constraints, since it applies to equality constraints (such
> as power balance) as well as inequality constraints.
>
> Unfortunately, there isn’t really an easy way to find which constraints
> “caused” an infeasible solution. My approach is usually to start by
> relaxing constraints until I get convergence. Then you can see which
> constraints violate the original limits and begin to gradually move the
> limits back toward their original values, watching which constraint
> multipliers grow large (indicating constraints which may conflict).
>
> Hope this helps,
>
> --
> Ray Zimmerman
> Senior Research Associate
> B30 Warren Hall, Cornell University, Ithaca, NY 14853  USA
> phone: (607) 255-9645
>
>
> On Sep 12, 2014, at 8:19 AM, ni le <[email protected]> wrote:
>
> Dear all,
>
> I'm running Optimal Power Flow using Matpower and I would like to know how
> can I know which constrains are not fulfilled if the OPF is not converged?
>
> In mpoption.m, there is an option to change the tolerance for constraint
> violation, i.e. OPF_VIOLATION, but I couldn't find this value used in the
> OPF program (e.g. in mips.m)? So how this value works in the OPF program?
>
> Thank you very much!
>
>
> Kind regards,
> ni le
>
>
>

Reply via email to