Prediction of Users’ Behaviour Patterns Impact on Energy Performance of a Social Housing in Cremona, Italy

Social housing is a crucial challenge of European cities, facing the pressing need to provide accommodation for an increasing and changing population with new family models. Functionality, adaptability and quality of living are fundamental topics in social housing. Economic constraints concur to shape the size of apartments and construction choices in the design phase. Further, energy efficiency is a key factor for long-term economic sustainability by reducing running cost in the operation phase. The availability of calibrated energy models is fundamental to perform effective energy management during building lifespan and the ability to investigate users’ behaviour by means of physical-statistical models is fundamental to improve current design and operation practices. The case study presented is a social housing located in Cremona, Italy, built by a Social Fund engaged in retrofit interventions on existing social housing and new construction. The variability of energy use, resulting from different occupancy, appliances, ventilation patterns and comfort settings has been investigated by means of simulations. Behavioural patterns are inherently dependent on age, number of components of the family, comfort preferences and activities (i.e. appliances and lighting use). Further, possible occupants’ aggregations are constrained by size and flexibility of the architectural layout of residential units (e.g. number of users, change in ability due to ageing, etc.). The results are presented in a graphical way (i.e. energy demand vs outdoor temperature), enabling statistical correlation among energy demand and physical parameters assumed in different simulation scenarios, which can be used for model calibration in the operation phase (for energy management purposes). The results show the large variability of energy performance in different scenarios and the large difference with respect to standard assumptions, highlighting the benefits of parametric behavioural simulation to obtain realistic data for techno-economic assessments.