This policy brief highlights two interconnected research initiatives focused on estimating energy consumption in buildings: the first examines the energy modeling of Italy's residential building stock, while the second focuses on analyzing and modeling real energy data for non-residential buildings.
The first contribution advances energy modeling for Italy’s residential sector to support decarbonization goals under the European Green Deal. It addresses the sector’s significant energy consumption and carbon footprint through a refined modeling approach.
The study builds on data from the 2013 ISTAT household energy survey, covering 20,000 households across Italy. The model integrates building archetypes, regional climate variability, energy simulation tools for heating and cooling demand. It also accounts for incomplete data and user behavior via probabilistic methods. The model results accurately replicate national energy consumption patterns: there is less than a 9% error compared to data from TERNA and IEA.
A scenario analysis is also implemented. The key findings reveal that retrofit measures and the adoption of renewable energy technologies, such as heat pumps, offer substantial energy savings and emissions reductions. For this reason, they should be considered for future policy development related to the Italian energy transition.
The second research investigates energy consumption patterns across a sample of University buildings, including schools, offices, laboratories, and recreational facilities. The analysis focuses on the building stock of the Alma Mater Studiorum University of Bologna, and it evaluates various energy vectors, such as electricity, natural gas, and district heating, identifying diverse end uses and their impact.
This work addresses two challenges: on the one hand, the multi-purpose nature of university buildings; on the other hand, the diversification of energy consumption end uses. The proposed method to estimate annual electrical and thermal consumption profiles on an hourly basis is based on categorizing days into typical types (e.g., summer and winter working days, weekends, holidays).
By identifying consumption drivers for each end use, normalized trends were derived using Fourier transforms and other analytical techniques. Key findings include the value of analyzing electricity consumption in different time slots to determine predominant building use and the potential to estimate consumption for different services in the absence of divisional meters. This work contributes to improving energy efficiency and sustainability in public buildings.
