In this study, a controller based on deep reinforcement learning was tested for a residential building equipped with a radiant heating system. In detail, a Soft Actor-Critic (SAC) algorithm was implemented to optimize the operation of the heating system while ensuring adequate levels of indoor temperature. A probabilistic window opening behavior model was implemented within the simulation framework in order to emulate the interaction of the occupants with the building. A sensitivity analysis on SAC hyperparameters was carried out to determine the best configuration that was then deployed in four different scenarios in order to analyze the adaptability of the controller to different boundary conditions. The performance of the reinforcement learning agent was evaluated against a baseline strategy which combines rule-based and climatic control. The developed agent was able to achieve a saving of heating energy provided to the building in the range between 2 and 6% while increasing temperature control performance up to 65% in the four scenarios investigated.

Energy Management of a Residential Heating System Through Deep Reinforcement Learning / Brandi, S.; Coraci, D.; Borello, D.; Capozzoli, A.. - STAMPA. - 263:(2022), pp. 329-339. (Intervento presentato al convegno 13th KES International Conference on Sustainability and Energy in Buildings, SEB 2021 nel 2021) [10.1007/978-981-16-6269-0_28].

Energy Management of a Residential Heating System Through Deep Reinforcement Learning

Brandi S.;Coraci D.;Capozzoli A.
2022

Abstract

In this study, a controller based on deep reinforcement learning was tested for a residential building equipped with a radiant heating system. In detail, a Soft Actor-Critic (SAC) algorithm was implemented to optimize the operation of the heating system while ensuring adequate levels of indoor temperature. A probabilistic window opening behavior model was implemented within the simulation framework in order to emulate the interaction of the occupants with the building. A sensitivity analysis on SAC hyperparameters was carried out to determine the best configuration that was then deployed in four different scenarios in order to analyze the adaptability of the controller to different boundary conditions. The performance of the reinforcement learning agent was evaluated against a baseline strategy which combines rule-based and climatic control. The developed agent was able to achieve a saving of heating energy provided to the building in the range between 2 and 6% while increasing temperature control performance up to 65% in the four scenarios investigated.
2022
978-981-16-6268-3
978-981-16-6269-0
File in questo prodotto:
File Dimensione Formato  
Energy Management of a Residential Heating System Through Deep Reinforcement Learning.pdf

Open Access dal 01/10/2022

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 712.63 kB
Formato Adobe PDF
712.63 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2938752