Updates for integration of the EV fleet

GMLC-1-4-2 / battery_interface

Implemenation of Device Models and their Battery Equivalent Interface

MIT License

5 stars 0 forks source link

Updates for integration of the EV fleet #61

Closed afernandezcanosa closed 5 years ago

afernandezcanosa commented 5 years ago

Some changes to make the integration of the EV fleet simpler.

afernandezcanosa commented 5 years ago

@hlngo @Hayden-Reeve @emayhorn I have added this PR to allow the artificial inertia service to pass the initial time stamp. This way, the user can see the ability of a particular device to respond to frequency deviations.

For example, at midnight, when all the vehicles are being charged, frequency deviations are well tracked:

discretized_freq_response_00_00

However, at 10 AM, when most of the sub fleets are at work or driving, the fleet won't be able to respond well:

discretized_freq_response_10_00

Hayden-Reeve commented 5 years ago

@afernandezcanosa ,

These look like great results and give a great indication of how we can show how diurnal cycles affect performance. Copying in @yliu250 so he is aware.

afernandezcanosa commented 5 years ago

@Hayden-Reeve Exactly, I guess that the opposite will be happening for PVs.

By the way, @yliu250 I was thinking about possible performance metrics of this service and I think that reporting a sensibility metric will be great. As we don't have a P_request for this service, something like the derivative of the service power with respect to the frequency divided by the fleet rating could be very useful. What do you think?

yliu250 commented 5 years ago

For now, I will probably implement service efficacy metrics, which calculate the integration of service power over time, divided by the total drive cycle energy.