search

ugr-sail / sinergym

Gym environment for building simulation and control using reinforcement learning
https://ugr-sail.github.io/sinergym/
MIT License
134 stars 37 forks source link

Noise improvement using Olstein-Uhlenberg processes #54

Closed AlejandroCN7 closed 3 years ago

AlejandroCN7 commented 3 years ago

In order to generate better stochastic environments, we could develop noise generation using Olstein-Uhlenberg processes

AlejandroCN7 commented 3 years ago

These are some results applying Olstein-Uhlenberg processes. Red line describe drybulb in epw with noise and blue line the same but in original epw file:

sigma=1.0 mu=0.0 tau=0.001 Captura de pantalla de 2021-07-02 10-18-15

sigma=5.0 mu=0.0 tau=0.01 Captura de pantalla de 2021-07-02 10-08-53

sigma=10.0 mu=0.0 tau=0.1 Captura de pantalla de 2021-07-02 10-21-20