Antares-Simulator is an Open Source power system simulator to quantify the adequacy or the economic performance of interconnected energy systems, at short or remote time horizons.
Context :
Many production means ara supplied with noises to avoid execution instability and force a choice between weekly optimization's equivalent solutions.
In general, there's a noise a given year and area or each thermal clusters of an area.
There are roughly 2 steps for every noise :
update : for each year, area / clusters of an area, an update of the related noise is done
use : then the noises are used to fill a field of the weekly optimization problem
Problem : lack of clarity and simplicity
Update : instead of being done in one place, this update can be done in 2 different places. For instance, for thermal noises, a partial update is done before running a batch of parallel year (see function computeRandomNumbers(...)), and the rest inside a MC year (see function PrepareRandomNumbers). And the same for hydro noises.
Use : we're filling the weekly problem with noises. Sometimes we do that inside a year, in function PrepareRandomNumbers (see for instance noises on un-supplied or spilled energy costs), and sometimes we do that in BuildThermalPartOfWeeklyProblem for thermal noises. Do we want it to be done like in different places depending on the production means ?
pet-mit
commented
1 month ago
Context : Many production means ara supplied with noises to avoid execution instability and force a choice between weekly optimization's equivalent solutions. In general, there's a noise a given year and area or each thermal clusters of an area. There are roughly 2 steps for every noise :
Problem : lack of clarity and simplicity