Unfortunately, since MOST is currently DC only, it cannot handle the reactive power dispatch. Carlos Murillo has done some work on an AC version, but I’m not sure it ever reached a usable state or has been updated to be compatible with the released version of MOST.
One approach is to simply implement your own multiperiod AC OPF using the strategy described in this post <https://www.mail-archive.com/[email protected]/msg00702.html>. Another possible approach would be to use MOST with DC network modeling to solve for the commitment status and tentative dispatch of the generators that can be used to impose fixed ramp constraints on a set of single-period AC OPF runs that provide the final active and reactive power dispatches. The idea is that you need some sort of multiperiod optimization to handle inter temporal constraints, such as ramping limits. For example, maybe you could use the active power dispatches from the MOST run for periods t–1 and t+1, along with the ramp rates, to define a feasible dispatch range (PMIN and PMAX) that you impose when solving the subsequent AC OPF for period t. This would hopefully result in solutions whose transitions from period to period are feasible, and whose per-period solutions are AC optimal within some restricted feasible dispatch range. I’m not sure whether the differences between AC and DC would still result in infeasibilities, but I suppose it could. Just an idea … Ray > On Feb 21, 2018, at 5:32 AM, Zenifar parekh <[email protected]> > wrote: > > Dear friends, > > Is it possible to solve multi-period optimal power flow problem for reactive > power using MOST? > > Basically, I need to derive optimal reactive power dispatch schedules for > each 15-minute time block of daily load curve. > > How can it be done by MOST? > > thanks in advance
