The example scenario PCM_FULL_YEAR starts at 1/1/2023 12:00 AM and does not have an initial state. I noticed that in the example solution for this scenario, for the first time step (1/1/2023 12:00 AM), demand served (P) is zero and PNS (demand not served) is non-zero for all EDEM objects. I also noticed that generation is zero for all generator externals at 1/1/2023 12:00 AM. I am wondering whether this is because:
(a) Setting initial state to NONE implies that all generators are offline at the start of the simulation. Generators need time to ramp up full output. I think this is probably not what is going on, as the FGEN.VERMONT_YANKEE unit has a MaxUpRampRate of 1 MW/ min and while FGEN.VERMONT_YANKEE.P is 0 MW at 1/1/2023 12:00 AM, it jumps up to 540 MW by 1/1/2023 1:00 AM. If this was a true ramp, it would violate this constraint.
(b) The dispatch doesn’t actually start until 1/1/2023 1:00 AM. StartTime is listed as 1/1/2023 12:00 AM, but this is the first time step in a scenario. Dispatch gets scheduled for the end of each time step, not the start.
Please let me know which is the correct explanation for this behavior - or if they are both wrong, and there’s another explanation!
Thank you for your question! I think there are 2 points that I can clear up here. If you still have follow up questions, please respond directly here and tag me so I know to respond!
An initial state of NONE does not actually imply that all generators are off prior to the beginning of the scenario. Rather, it implies that their status is “undetermined” prior to the beginning of the scenario. This means they are free to dispatch to any level at the first timestep, and after that ramp rates are enforced. Furthermore, this does not qualify as a “start-up” or a “shut-down” so generators are not subject to these constraints until after a start-up or shut-down occurs during the scenario.
1a) You can specify the status of a generator prior to the beginning of the scenario using the events “ONINI”, “OFFINI”, or “PINI”. You can also specify the status of every object at the beginning of a scenario by referring to a state (*.econ) file. No such events or initial conditions are specified for this example scenario in the ENET39 dataset.
For any scenario in SAInt, the “StartTime” represents the initial conditions for the scenario. The first timestep in this scenario is identified by the timestamp “1/1/2023 1:00 AM”. Please don’t think of it as dispatch being scheduled for the “end” of each time step. Rather, the time point at the end of each time step represents the full period of time for that time step. E.g., the time point of 1 AM on the 1st of January represents the period between 12 AM and 1 AM on that day.
Please let me know if you have any further questions.