Why are the action and observation always out of sync? [Question]

ugr-sail / sinergym

Gym environment for building simulation and control using reinforcement learning

https://ugr-sail.github.io/sinergym/

MIT License

131 stars 35 forks source link

Why are the action and observation always out of sync? [Question] #404

Closed jingjijiji closed 7 months ago

jingjijiji commented 7 months ago

Question ❓

when i try the bacic_example.ipynp, i found that actions do not take effect immediately, but rather after two steps. Why is that?

Additional context

Add any other context about the question here.

Checklist

[x] I have read the documentation (required)
[x] I have checked that there is no similar issue in the repo (required)

:pencil: Please, don't forget to include more labels besides question if it is necessary.

AlejandroCN7 commented 7 months ago

Hi @jingjijiji !

This is due to the functioning of the EnergyPlus API which is what Sinergym uses as middleware. It has a delay from the time the action is sent until the change occurs in the building, usually two steps as you mentioned. At the moment, I have no more information about it.