fleetrl.fleet_env.config package

Submodules

fleetrl.fleet_env.config.ev_config module

class fleetrl.fleet_env.config.ev_config.EvConfig(env_config)[source]

Bases: object

The EV config sets some default parameters regarding the vehicles, charging infrastructure, efficiencies, etc.

fleetrl.fleet_env.config.score_config module

class fleetrl.fleet_env.config.score_config.ScoreConfig(env_config)[source]

Bases: object

The Score Config sets coefficients to calculate the reward function - Multipliers: price_multiplier, penalty_invalid_action, penalty_overcharging, penalty_overloading - Changing the multipliers to 0 leads to ignoring this aspect of the reward function - The SOC violation and overloading are calculated using sigmoid functions

overloading_penalty(rel_loading)[source]

static soc_violation_penalty(missing_soc)[source]

fleetrl.fleet_env.config.time_config module

class fleetrl.fleet_env.config.time_config.TimeConfig(env_config)[source]

Bases: object

The Time config sets parameters for the episode and MDP dynamics. - end_cutoff: Default at 60. The last two months are not regarding in the training observations - The last two months are used as a validation set during training