sinergym.envs.eplus_env.EplusEnv

class sinergym.envs.eplus_env.EplusEnv(*args: Any, **kwargs: Any)

Environment with EnergyPlus simulator.

__init__(idf_file, weather_file, variables_file, spaces_file, env_name='eplus-env-v1', discrete_actions=True, weather_variability=None, reward=<sinergym.utils.rewards.LinearReward object>)

Environment with EnergyPlus simulator.

Parameters:

• idf_file (str) – Name of the IDF file with the building definition.

• weather_file (str) – Name of the EPW file for weather conditions.

• variables_file (str) – Variables defined in environment to be observation and action (see sinergym/data/variables/ for examples).

• spaces_file (str) – Action and observation space defined in a xml (see sinergym/data/variables/ for examples).

• env_name – Env name used for working directory generation.

• discrete_actions (bool, optional) – Whether the actions are discrete (True) or continuous (False). Defaults to True.

• weather_variability (tuple, optional) – Tuple with sigma, mu and tao of the Ornstein-Uhlenbeck process to be applied to weather data. Defaults to None.

• reward (Reward instance) – Reward function instance used for agent feedback. Defaults to LinearReward.

Methods

__init__(idf_file, weather_file, ...[, ...])	Environment with EnergyPlus simulator.
close()	End simulation.
render([mode])	Environment rendering.
reset()	Reset the environment.
step(action)	Sends action to the environment.

Attributes

metadata

close()

End simulation.

metadata = {'render.modes': ['human']}

render(mode='human')

Environment rendering.

reset()

Reset the environment.

Returns:

Current observation.

Return type:

np.array

step(action)

Sends action to the environment.

Parameters:

action (int or np.array) – Action selected by the agent.

Returns:

Observation for next timestep. float: Reward obtained. bool: Whether the episode has ended or not. dict: A dictionary with extra information.

Return type:

np.array