S
Stefano Pizzuti
Researcher at ENEA
Publications - 57
Citations - 841
Stefano Pizzuti is an academic researcher from ENEA. The author has contributed to research in topics: Artificial neural network & Evolutionary algorithm. The author has an hindex of 13, co-authored 57 publications receiving 689 citations. Previous affiliations of Stefano Pizzuti include Environment Agency & Roma Tre University.
Papers
More filters
Journal ArticleDOI
Urban traffic flow forecasting through statistical and neural network bagging ensemble hybrid modeling
TL;DR: A hybrid modeling approach which combines Artificial Neural Networks and a simple statistical approach in order to provide a one hour forecast of urban traffic flow rates is shown.
Journal ArticleDOI
Adaptive Street Lighting Predictive Control
TL;DR: Experimental results provided by a real life testbed showed that the proposed strategy based on the given traffic forecasts and on the dynamical street class downgrade allowed by the law has high potential energy savings without affecting safety.
Journal ArticleDOI
Data-driven models for short-term thermal behaviour prediction in real buildings
Francesco Ferracuti,Alessandro Fonti,Lucio Ciabattoni,Stefano Pizzuti,Alessia Arteconi,Lieve Helsen,Gabriele Comodi +6 more
TL;DR: In this article, the authors compared three data-driven models for short-term thermal behavior prediction in a real building, part of a living smart district connected to a thermal network, and demonstrated that all the models investigated can also be proposed as a powerful tool to detect some typologies of occupant bad behaviours and to predict the shortterm flexibility of the building for demand response (DR) applications.
Journal ArticleDOI
The role of data sample size and dimensionality in neural network based forecasting of building heating related variables
Martin Macas,Fabio Moretti,Alessandro Fonti,Andrea Giantomassi,Gabriele Comodi,Mauro Annunziato,Stefano Pizzuti,Alfredo Capra +7 more
TL;DR: In this article, an early stopping mechanism is used for small training data, because it reliably overcomes overfitting problems, and two strategies of selection of suitable input variables are demonstrated.
Journal ArticleDOI
Smart street lighting management
TL;DR: The study shows that with the proposed approach, it is possible to save up to 50 % of energy compared to no regulation systems.