Vlad Iftime Kiflin PPBC #95
Conversation
…C control type to work only with TUNES RM (for now). Signed-off-by: Victor Garcia Reolid <[email protected]>
Signed-off-by: Victor Garcia Reolid <[email protected]>
…TUNES/send-system-description-to-fm # Conflicts: # README.rst # src/flexmeasures_client/client.py
…TUNES/send-system-description-to-fm # Conflicts: # setup.cfg
…ption-to-fm' into feature/TUNES/send-system-description-to-fm
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
VladIftime
force-pushed
the
VladIftime-Kiflin-PPBC
branch
2 times, most recently
from
d607033 to
2c373bd
Compare
Flix6x
changed the base branch from
main
to
feature/TUNES/send-system-description-to-fm
VladIftime
force-pushed
the
VladIftime-Kiflin-PPBC
branch
from
86cb3d4 to
b7694ff
Compare
…ftime-Kiflin-PPBC Signed-off-by: VladIftime <[email protected]>
Signed-off-by: Vlad Iftime <[email protected]>
VladIftime
force-pushed
the
VladIftime-Kiflin-PPBC
branch
from
0e39df6 to
c8d7e26
Compare
…TUNES/send-system-description-to-fm # Conflicts: # README.rst
…system-description-to-fm' into VladIftime-Kiflin-PPBC # Conflicts: # README.rst # src/flexmeasures_client/client.py # tests/test_client.py
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
…ation mode is set to that of the Tarnoc Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
| async def send_schedule_instruction(self, instruction: PPBCScheduleInstruction): | ||
| await self._fm_client.post_schedule( | ||
| self._power_sensor_id, | ||
| start=self.now(), | ||
| values=instruction.power_values, | ||
| unit="MW", | ||
| duration=self._schedule_duration, | ||
| price_sensor_id=self._price_sensor_id, | ||
| price_values=instruction.price_values, | ||
| price_unit="EUR/MWh", | ||
| valid_from=self.now() + self._valid_from_shift, | ||
| ) |
There was a problem hiding this comment.
This function isn't called anywhere yet. It looks like a placeholder for future work. There's also no client.post_schedule. I'm adding a todo to that extent.
Content-wise, I think it confuses sending a PPBCScheduleInstruction to the RM with asking the FlexMeasures server for a schedule.
There was a problem hiding this comment.
I saw the TODO's comments and I agree. I do think that before I can do a proper implementation for trigger_schedule I would need the new scheduler to be implemented so that I can figure out what kind of flex_model I need to pass. Is that right?
There was a problem hiding this comment.
I'm guessing PPBC should map to the flex-model for our ProcessScheduler. This function should then just call the same trigger_and_get_schedule method.
FlexMeasures currently does not support choosing which scheduler to use at the same time as triggering a schedule (not explicitly in the trigger message nor implicitly by checking what type of flex-model is given). Instead, I believe the scheduler is picked based on a sensor attribute (custom-scheduler) or possibly based on the asset type of the asset that the sensor belongs to. That means we should either make it possible to select a specific scheduler in the API call that triggers a schedule, or we have the flexmeasures-client set the dedicated sensor attribute when the PPBC control type is selected, and set it back to the StorageScheduler when the FRBC control type is selected. I prefer the latter as a short-term solution.
Can you create an issue where we can discuss the mapping PPBC <-> ProcessScheduler?
There was a problem hiding this comment.
Why get rid of the S2Message wrapper that adds the dt attribute (for doing simulations)?
tests/test_s2_models.py
Outdated
| "provides_power_measurement_types": ["ELECTRIC.POWER.3_PHASE_SYMMETRIC"], | ||
| "message_type": "ResourceManagerDetails", | ||
| }, | ||
| "metadata": {"dt": "2023-01-01T00:00:00"}, |
There was a problem hiding this comment.
In my opinion, these should be forced to be timezone aware datetimes (and following the ISO 8601 standard), e.g. "2023-01-01T00:00:00+00".
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
Signed-off-by: F.N. Claessen <[email protected]>
…system-description-to-fm' into VladIftime-Kiflin-PPBC
…tarted implemenation of trigger_schedule for PPBC simple
Added the skeleton for PPBC implementation into flex measures