Python Configuration

The config is supposed to simplify the user’s interaction with the package by providing a central place to set all parameters. By default, the config file has to contain three sections:

[General]
[FMU]
[Reinforcement Learning]

The parameters within the section will be converted to their respective types and are available within the environment class.

Required Parameters

The values are examples to show the required type.

Name	Value	Description
FMU_path	path/to/fmu	relative or absolute path to the file
stop_time	1	when the time is reached, the simulation ends [s]
dt	0.5	fixed timestep of the simulation [s]

Optional Parameters

Name	Value	Description
action_interval	1.5	every 1.5 seconds the agent can take an action [s]
reset_inputs	True	if this is true the inputs are reset to 0 between each episode
start_time	1	specify the start time of the FMU simulation [s]

Additional parameters will also be available as self.parameter.

Example Config

[General]
[FMU]
FMU_path = 00-Simulink_Linux.fmu
stop_time =  1
dt = 0.5

[Reinforcement Learning]