The MPC capability provided by
REopt is essentially the optimal sizing and dispatch capability that
REopt is commonly used for, but with the sizing problem removed. Also, the MPC model can be built for an arbitrary time length, or "horizon" (whereas a
REopt model always includes an entire calendar year). The MPC model also requires the user to provide load and resource forecasts as inputs (whereas the typical
REopt model allows one to use built-in load profiles as well as other API's such as PVWatts for the solar resource).
using REopt, JuMP, Cbc model = Model(Cbc.Optimizer) results = run_mpc(model, "./test/scenarios/mpc.json")
See mpc.json for details on the Scenario.