sustaindc_env module

class sustaindc_env.EnvConfig(raw_config)[source]

Bases: dict

DEFAULT_CONFIG = {'actions_are_logits': False, 'agents': ['agent_ls', 'agent_dc', 'agent_bat'], 'bat_reward': 'default_bat_reward', 'cintensity_file': 'NYIS_NG_&_avgCI.csv', 'datacenter_capacity_mw': 1, 'days_per_episode': 30, 'dc_config_file': 'dc_config.json', 'dc_reward': 'default_dc_reward', 'evaluation': False, 'flexible_load': 0.1, 'individual_reward_weight': 0.8, 'location': 'ny', 'ls_reward': 'default_ls_reward', 'max_bat_cap_Mw': 2, 'timezone_shift': 0, 'weather_file': 'USA_NY_New.York-Kennedy.epw', 'workload_file': 'Alibaba_CPU_Data_Hourly_1.csv'}
class sustaindc_env.SustainDC(env_config)[source]

Bases: Env

calculate_reward(params)[source]

Calculate the individual reward for each agent.

Parameters:

params (dict) – Dictionary of parameters to calculate the reward.

Returns:

Individual reward for the load shifting agent. dc_reward (float): Individual reward for the data center agent. bat_reward (float): Individual reward for the battery agent.

Return type:

ls_reward (float)

close()[source]

Close the environment.

get_avail_actions()[source]

Get the available actions for the agents.

Returns:

List of available actions for each agent.

Return type:

list

get_avail_agent_actions(agent_id)[source]

Get the available actions for a specific agent.

Parameters:

agent_id (int) – Agent ID.

Returns:

List of available actions for the agent.

Return type:

list

render()[source]

Render the environment.

reset()[source]

Reset the environment.

Parameters:
  • seed (int, optional) – Random seed.

  • options (dict, optional) – Environment options.

Returns:

Dictionary of states. infos (dict): Dictionary of infos.

Return type:

states (dict)

seed(seed=None)[source]

Set the random seed for the environment.

Parameters:

seed (int, optional) – Random seed.

state()[source]

Get the state of the environment.

Returns:

State of the environment.

Return type:

np.ndarray

step(action_dict)[source]

Step the environment.

Parameters:

action_dict – Dictionary of actions of each agent defined in self.agents.

Returns:

Dictionary of observations/states. rews (dict): Dictionary of rewards. terminated (dict): Dictionary of terminated flags. truncated (dict): Dictionary of truncated flags. infos (dict): Dictionary of infos.

Return type:

obs (dict)