Towards adaptive building automation systems, simulating occupant behaviors by means of an agent based approach

The recent international directives have introduced the "Nearly Zero Energy Building" standard in response to the high impact of buildings on energy consumptions and greenhouse gas emissions. Most of the innovative strategies and technology solutions for reducing energy consumption in buildings are focused on energy efficiency of building-HVAC system. The interaction between occupants and the building-HVAC system is completely neglected, especially regarding the operating modes by occupants. In most cases, it causes a significant difference between the expected energy consumptions and the real ones during the exercise phase. Especially in buildings with high levels of thermal insulation and airtightness, occupant behavior may have a significant influence on energy consumption and on indoor comfort. In this context, the aim of this research work is to evaluate the occupants' behavior impact on energy consumption and thermal comfort, in order to design building automation systems adaptable to occupants behaviors. In particular, unlike the common deterministic approach to simulate occupant behavior in buildings, this work presents a more realistic approach to simulate occupant behavior by means of a stochastic and agent-based model (ABM). The case study regards the residential building context, where behavioral logic of occupants relating to the opening/closing of the windows and shading and to the activation of the cooling system in summer are implemented by TRNSYS software. The comparison between models where occupants are simulated in deterministic way and by means of ABM allows to evaluate the significant impacts of occupants on building performance